WO2023190096A1 - 虫同定用画像データの生成方法及び虫同定システム - Google Patents
虫同定用画像データの生成方法及び虫同定システム Download PDFInfo
- Publication number
- WO2023190096A1 WO2023190096A1 PCT/JP2023/011675 JP2023011675W WO2023190096A1 WO 2023190096 A1 WO2023190096 A1 WO 2023190096A1 JP 2023011675 W JP2023011675 W JP 2023011675W WO 2023190096 A1 WO2023190096 A1 WO 2023190096A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- insect
- data
- image data
- information
- point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Ceased
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/145—Illumination specially adapted for pattern recognition, e.g. using gratings
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/22—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition
- G06V10/225—Image preprocessing by selection of a specific region containing or referencing a pattern; Locating or processing of specific regions to guide the detection or recognition based on a marking or identifier characterising the area
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
- G06V10/757—Matching configurations of points or features
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/774—Generating sets of training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01M—CATCHING, TRAPPING OR SCARING OF ANIMALS; APPARATUS FOR THE DESTRUCTION OF NOXIOUS ANIMALS OR NOXIOUS PLANTS
- A01M1/00—Stationary means for catching or killing insects
- A01M1/02—Stationary means for catching or killing insects with devices or substances, e.g. food, pheronones attracting the insects
- A01M1/026—Stationary means for catching or killing insects with devices or substances, e.g. food, pheronones attracting the insects combined with devices for monitoring insect presence, e.g. termites
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01M—CATCHING, TRAPPING OR SCARING OF ANIMALS; APPARATUS FOR THE DESTRUCTION OF NOXIOUS ANIMALS OR NOXIOUS PLANTS
- A01M1/00—Stationary means for catching or killing insects
- A01M1/14—Catching by adhesive surfaces
-
- A—HUMAN NECESSITIES
- A01—AGRICULTURE; FORESTRY; ANIMAL HUSBANDRY; HUNTING; TRAPPING; FISHING
- A01M—CATCHING, TRAPPING OR SCARING OF ANIMALS; APPARATUS FOR THE DESTRUCTION OF NOXIOUS ANIMALS OR NOXIOUS PLANTS
- A01M2200/00—Kind of animal
- A01M2200/01—Insects
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING OR CALCULATING; COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A50/00—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE in human health protection, e.g. against extreme weather
- Y02A50/30—Against vector-borne diseases, e.g. mosquito-borne, fly-borne, tick-borne or waterborne diseases whose impact is exacerbated by climate change
Definitions
- the present invention relates to a method for generating image data for insect identification and an insect identification system.
- the captured insect identification system includes an image reading unit that reads an image of an adhesive sheet to which an insect is attached, and an analysis device that analyzes the read image and identifies the captured insect.
- the analysis device includes a preprocessing means that performs image processing on the loaded image to distinguish between the insect area, which is the area in which the insect is photographed, and the background, and extracts insect merkmar data from the preprocessed image.
- a data storage means for storing standard Merkumar data obtained by standardizing Merkumar of insect morphological characteristics; and Merkumar data extracted by the feature extraction means and standard Merkumar data stored in the data storage means; identification means for comparing and identifying insects.
- the captured insects take various postures, so for example, when the insect is bent and attached to the adhesive sheet, Features could not be extracted properly, leading to incorrect recognition and lower identification accuracy.
- an object of the present invention is to provide a method for generating image data for insect identification and an insect identification system that can improve the accuracy of identifying insects included in image data.
- the method of generating image data for insect identification of the present invention includes an insect extraction step of extracting insect data regarding one insect from unprocessed image data in which insects are included in the image, and at least one of the extracted insect data. a point information adding step of specifying three points and adding point information data to the insect data; and an order information adding step of adding order information regarding the order to the point information data of the at least three points specified by the point information adding step. and has.
- the insect identification system of the present invention includes: an image data acquisition unit that acquires target image data in which an insect to be identified is included in the image; an insect identification unit that identifies the type of insect included in the target image data;
- the insect identification unit is configured to identify the type of insect included in the target image data acquired by the image data acquisition unit using a learned model constructed by machine learning using teacher data,
- the trained model includes teacher image data in which an insect is included in the image, point information data that is assigned to the insect data by specifying at least three points among the insect data related to insects included in the teacher image data, and , posture information including order information regarding the order given to the point information data, and type information regarding the type of insect included in the teacher image data.
- FIG. 2 is a schematic diagram showing a photographing device for photographing unprocessed image data in a method of generating images for insect identification according to an embodiment of the present invention. It is a flowchart which shows the process in the generation method of the image for a contagious insect identification. It is a flowchart which shows the process of the part extraction process in the generation method of the image for a contagious insect identification.
- FIG. 6 is a diagram showing an image corresponding to unprocessed image data to which frame information has been added in a method of generating images for identification of the same insect.
- FIG. 3 is a diagram showing insects corresponding to insect data to which point information data and order information have been added in a method for generating images for identifying the same insect, and showing insects corresponding to insect data whose parts boundaries can be recognized.
- FIG. 7 is a diagram showing insects corresponding to insect data to which point information data and order information are added in the method for generating images for identifying the same insect, and showing insects corresponding to insect data in which the boundaries of parts are unrecognizable.
- FIG. 1 is a diagram showing an overview of an insect identification system according to an embodiment of the present invention.
- FIG. 1 is a block diagram showing a congener identification system.
- FIG. 2 is a block diagram showing a pest control system that performs countermeasure control. It is a flowchart which shows the process of the insect increase determination, generation source, and treatment method determination process in the pest control system that performs countermeasure control. It is a flowchart which shows the process of the determination process of the chemical
- FIG. 1 is a block diagram showing a pest control system that performs preventive control. It is a flowchart which shows the flow of processing in a pest control system that performs preventive control. This is a data table used for determining insect growth environment information in a pest control system that performs preventive control.
- FIGS. 1 to 8 A method for generating insect identification image data and an insect identification system 6 according to an embodiment of the present invention will be described with reference to FIGS. 1 to 8.
- the unprocessed image data processed by the insect identification image data generation method of this embodiment will be explained with reference to FIG.
- the unprocessed image data is generated by an imaging device 3 capable of capturing an image of the capture surface 2a on which the insect 1 is captured in the captured sample 2 having the capture surface 2a on which the insect 1 can be captured.
- the captured sample 2 of this embodiment is a sample in which the insects 1 can be attached to the capture surface 2a, and a plurality of insects 1 are captured in a state that they are attached to the capture surface 2a.
- 1 is the paper that caught the insect.
- "insect" includes all small animals other than mammals, birds, and seafood, and in this embodiment refers to arthropods (particularly insects and spiders).
- the capture surface 2a has a flat shape extending in one direction (in this embodiment, the direction penetrating the plane of FIG. 1). Further, the capture surface 2a is a plain color, and in this embodiment, the entire surface is yellow, and is a surface without ruled lines or the like.
- a plurality of insects 1 are attached to the capture surface 2a of the captured sample 2.
- the captured sample 2 of this embodiment includes a sample created by sprinkling the dead insect 1 on the capture surface 2a and sticking it, or a sample created by attracting the insect 1 with light or smell and making it stick to the capture surface 2a. Samples may be employed, but are not limited to these.
- the imaging device 3 is a device that images the captured sample 2 and generates unprocessed image data.
- the imaging device 3 includes a box-shaped casing 31 that can block external light, and is provided inside the casing 31, with the capturing surface 2a extending in one direction (in this embodiment, , a holding means (not shown) capable of holding the captured sample 2 in a state extending in the direction penetrating the plane of the paper in FIG. an irradiation device 33 provided inside the casing 31 and capable of irradiating light onto the capture surface 2a of the captured sample 2; and an imaging processing unit (not shown) that generates the image.
- the imaging means 32 includes a camera 34 arranged at a distance from the capture surface 2a so as to image the capture surface 2a of the captured sample 2 held by the holding means, and a camera 34 arranged apart from the capture surface 2a of the captured sample 2 held by the holding means.
- a traveling means (not shown) for causing the camera 34 to travel along the surface direction of the capture surface 2a is provided.
- the camera 34 is movable (traveling) along the surface direction of the capture surface 2a by a traveling means, and while moving in the surface direction of the capture surface 2a, the camera 34 captures the capture surface 2a at a plurality of positions in the surface direction of the capture surface 2a. Imaging is possible.
- the camera 34 of this embodiment is arranged at a position facing the capture surface 2a. Further, the camera 34 has a focal length such that only a portion of the capture surface 2a facing the camera 34 is in focus.
- the irradiation device 33 is configured to irradiate light onto the capture surface 2a from both sides of the travel area where the camera 34 travels. Specifically, the irradiation device 33 has illumination sections 35 on one side and the other side across the travel area, and each illumination section 35 irradiates light onto the capture surface 2a. Moreover, the illumination part 35 is provided so as to irradiate light from a position approximately 45 degrees upward from the center of the capture surface 2a in the width direction (direction perpendicular to the surface direction).
- the irradiation device 33 of this embodiment is configured such that a plurality of illumination units 35 are arranged in line along the running area, and can irradiate the entire area of the running area in one direction with light. Moreover, the irradiation device 33 of this embodiment irradiates the capture surface 2a with white light.
- the imaging processing unit connects a plurality of images captured by the imaging means 32 to generate unprocessed image data. Specifically, the image capture processing unit generates unprocessed image data by processing a plurality of images captured by the camera 34 and joining together in-focus portions of each image. In this embodiment, the image capture processing unit generates unprocessed image data by joining together images of the portion of the capture surface 2a facing the camera 34 for a plurality of images captured by the camera 34.
- the unprocessed image data generated as described above is, for example, data representing an image of the insect 1 placed on the capture surface 2a, as shown in FIG.
- the pre-processing image data of this embodiment is image data representing an image in which a plurality of insects 1 are photographed on the background of the capture surface 2a, and the pre-processing image data includes insect data D regarding the insects 1. Contains multiple. Insect data D is image data of a portion of the unprocessed image data that indicates insect 1.
- the method for generating image data for insect identification is a method for generating image data for insect identification as teacher data used in machine learning.
- the method for generating image data for insect identification processes unprocessed image data in which insect 1 is included in the image as described above to generate image data for insect identification. It's a method. Specifically, the method for generating insect identification image data includes an insect extraction step S1 of extracting insect data D regarding insect 1 from unprocessed image data, and extracting the parts of insect 1 included in the extracted insect data D. A part extraction step S2, a point information adding step S3 of adding point information data 4 to the extracted insect data D, and an order information adding step S4 of adding order information 5 regarding the order to the point information data 4. , is provided.
- each of the above steps is executed for all the insect data D, and then the next step is executed. Furthermore, in this embodiment, each step is performed by a person.
- the insect extraction step S1 is a step of executing an insect discovery step S11 for discovering the insect 1 included in the image corresponding to the unprocessed image data, and an insect identification step S12 for identifying the discovered insect 1. That is, in the insect extraction step S1, insect 1 is discovered in the insect discovery step S11, and in the insect identification step S12, data to which information for identifying the discovered insect 1 is added is extracted as insect data D.
- the insect finding step S11 is a step of processing the unprocessed image data and discovering the insect 1 included in the image corresponding to the unprocessed image data.
- the insect finding step S11 of this embodiment is a step of searching for a part having the characteristics of insect 1 from the image corresponding to the unprocessed image data, and specifically, a part of a color different from the background is an outline of a predetermined shape. , the part surrounded by the outline is determined to be one insect 1.
- the insect identification step S12 includes a frame information adding step, which is a step of adding frame information, which is information about the frame F surrounding the insect 1 found in the insect finding step S11, and a frame information adding step.
- This is a step of executing an extraction step of extracting insect data based on the assigned frame information, and a type information adding step of adding type information regarding the type of the discovered insect 1. That is, in the insect identification step S12 of the present embodiment, information regarding the range of insect 1 and the type of insect 1 included in the image corresponding to the unprocessed image data is specified, and information regarding the insect 1 related to the identified insect 1 is identified. Extract data D.
- the frame information adding step is a step of specifying the range occupied by the discovered insect 1 in the unprocessed image data, and adding frame information regarding the frame F surrounding the range occupied by the insect 1.
- the frame information adding step is a step of specifying the outer edge of the insect 1 discovered in the bug finding step S11, and adding information regarding the frame F that surrounds the outer edge.
- the frame F corresponding to the frame information provided in this embodiment is rectangular, particularly rectangular, and for example, the coordinates of two diagonally located points of the rectangular frame F are provided as the frame information. do.
- the extraction step is a step of extracting insect data D based on the frame information. Specifically, in the extraction step, data in which the range in which the insect 1 found in the insect discovery step S11 is located is surrounded by a frame F is extracted as insect data D. That is, the insect data D is image data extracted from the unprocessed image data by enclosing the range where the insect 1 is located with a frame F.
- the type information provision step includes the type of the insect 1 discovered (specifically, the type in a classification necessary to identify the source of the insect 1, which will be described later). This is a step of specifying the family (Family) in the classification class of and adding information about the type of insect 1 to insect data D.
- the information regarding the type provided is information that allows the type of insect 1 to be specified, and is, for example, the type name of insect 1 or an identifier regarding the type of insect 1.
- the type information adding step information regarding the type name of the insect 1 surrounded by the frame F corresponding to the frame information is added to the insect data D.
- the region extraction step S2 is a step of extracting the region of the insect 1 corresponding to each extracted insect data D. Further, in the part extraction step S2 of the present embodiment, it is determined whether or not the part can be specified for the insect 1 corresponding to each insect data D, and if the part can be specified, the insect 1 corresponding to each insect data D is determined. Identify and extract the part of insect 1. Specifically, the part extraction step S2 includes a selection step S21 in which the insect 1 corresponding to the insect data D to be processed is selected, and a part of the insect 1 corresponding to the insect data D selected in the selection step S21 can be specified.
- a repetition step S25 is executed in which each process in the part extraction step S2 is repeated. .
- the insect 1 includes a head 13 , a body 11 having a body 14 connected to the head 13 and including at least an abdomen 16 , and an appendage 12 extending outward from the body 11 .
- the appendage 12 includes, for example, antennae, legs, wings, and the like.
- the structure of the body 11 differs depending on the insect 1.
- the insect body 11 includes a head 13, a thorax 15 and an abdomen 16 as a torso 14.
- the spider body 11 includes a cephalothorax as a head 13 and an abdomen 16 as a trunk 14.
- the parts of the insect 1 will be explained as referring to each part in the body main body 11 of the insect 1.
- the selection step S21 is a step of selecting one insect 1 from among the insects 1 corresponding to the plurality of extracted insect data D as a processing target. Furthermore, in the selection step S21, the insect 1 corresponding to the unprocessed insect data D that has not been processed in the part extraction step S2 is selected.
- the feasibility determination step S22 is a step of determining whether the part of the insect 1 corresponding to the selected insect data D can be specified. In the feasibility determination step S22 of the present embodiment, it is determined whether each part appears to a distinguishable degree in the image of the insect 1 corresponding to the selected insect data D, and each part appears to a distinguishable degree. In this case, it is determined that the part of the insect 1 corresponding to the selected insect data D can be specified. In the feasibility determination step S22 of this embodiment, it is determined that the parts can be distinguished if at least the head 13 and the torso 14 can be distinguished. For example, as shown in FIG.
- the part cannot be identified when the part 1 cannot be determined.
- the part of the insect 1 can be identified as long as the body 11 can be recognized to the extent that the part of the insect 1 can be determined. It is determined that
- the identifying step S23 is a step of identifying and extracting each part of the insect 1 for which it was determined in the availability determining step S22 that the part of the insect 1 corresponding to the insect data D can be identified. Specifically, the identifying step S23 identifies and extracts the range in which each part is located in the insect 1 image data corresponding to the insect data D. For example, as shown in FIG. 5A, in the identification step S23, if the entire body 11 of the insect 1 corresponding to the insect data D can be recognized and the boundaries of the head 13, thorax 15, and abdomen 16 can be recognized, , of the body 11 of the insect 1, the range where the head 13 is located, the range where the thorax 15 is located, and the range where the abdomen 16 is located are identified and extracted.
- the boundary between the head 13 and the thorax 15 of the insect 1 corresponding to the insect data D can be recognized, but the boundary between the thorax 15 and the abdomen 16 cannot be recognized, or when the torso 14 is
- an insect 1 for example, a spider
- the range where the head 13 is located and the range where the body 14 is located are determined. At least two or more parts are specified and extracted, and information corresponding to each part is given to each extracted range.
- the non-designation processing step S24 is a step of specifying that the part of the insect 1 corresponding to the insect data D for which it was determined that the part cannot be determined in the feasibility determination step S22 is not to be identified.
- the insect 1 corresponding to the insect data D determined to be unable to identify its part is given information indicating that its part cannot be determined, for example, as shown in FIG. 5B.
- the repeating step S25 it is determined whether there is an insect 1 corresponding to the unprocessed insect data D that has not been processed in the part extraction step S2, and if there is an unprocessed insect 1, each process in the part extraction step S2 is performed. It is a process of repeating. Specifically, in the repeating step S25, if the plurality of insect data D included in the unprocessed image data include insect data D for which the processing in the specific step S23 or the processing in the unspecified processing step S24 has not been completed. determines that there is insect 1 corresponding to unprocessed insect data D, and proceeds to the selection step S21. Further, in the repeating step S25, if the processing in the specifying step S23 or the processing in the unspecified processing step S24 has been completed for all the insect data D included in the unprocessed image data, the part extraction step S2 is ended. .
- the point information adding step S3 is a step of specifying at least three points and adding point information data 4 to the insect data D.
- the point information adding step S3 three points on the surface of the insect 1 corresponding to the insect data D are specified and point information data 4 is added.
- Point information data 4 is assigned by specifying three points spaced apart in the direction (direction connecting the head and belly of insect 1). That is, the point information adding step S3 is a step of adding at least three points of point information data 4 within the range where the insect 1 corresponding to the insect data D is located in the unprocessed image data, and specifically, Point information data 4 is given to at least three points within the range where each part of the insect 1 extracted in the part extraction step S2 is located.
- the point information data 4 is set so that three points correspond to two or more of the plurality of parts constituting the body 11 of the insect 1 corresponding to the insect data D.
- point information data 4 is provided so that three points correspond to the head 13 and the torso 14. As shown in FIGS.
- the point information provision step S3 of this embodiment for the insect 1 corresponding to each insect data D, at least a position corresponding to the head 13 of the insect 1 and a position corresponding to the body 14 of the insect 1 Specifically, a head corresponding point 41 corresponding to the head 13 of insect 1, an abdomen corresponding point 42 corresponding to the abdomen 16 of insect 1, and a head Three points, an intermediate point 43 located between the corresponding point 41 and the abdomen corresponding point 42, are designated and point information data 4 is assigned.
- point information data 4 is provided based on the part of the insect 1 corresponding to the insect data D, so in the part extraction step S2, the insect data from which the part has been extracted is Point information data 4 is assigned to insect data D, but point information data 4 is not assigned to insect data D for which it is determined that the body part cannot be determined in the part extraction step S2.
- the point information data 4 is supplementary information indicating the coordinates in the insect data D, but the configuration is not limited to this, and the insect image as the insect data D may be overwritten with points. It is also possible to use a configuration in which the overwritten points are handled as point information data 4.
- the head corresponding point 41 is a point specified at the tip of the head 13 (the end on the head side in the overall length direction of the body 11), and is also specified in the width direction of the body 11 (the insect corresponding to the insect data D). This is a point designated approximately at the center of the image (in a direction perpendicular to the overall length direction of the body 11). That is, the head corresponding point 41 is a point designated at the tip of the body 11 of the insect 1 and approximately at the center of the body 11 in the width direction.
- the abdomen corresponding point 42 is a point specified at the rear end of the abdomen 16 (the end on the ventral side in the overall length direction of the body body 11), and is also a point specified at approximately the center in the width direction of the body body 11. be. That is, the abdomen corresponding point 42 is a point designated at the rear end of the body 14 (the rear end of the body 11 of the insect 1) and approximately at the center of the body 11 in the width direction.
- the intermediate point 43 is a point designated in the middle of the body 11 of the insect 1 in the overall length direction. Further, the midpoint 43 is a point designated approximately at the center in the width direction of the insect 1, and in this embodiment, is a point located on the body 14. Furthermore, the intermediate point 43 is a point located at a feature point in the body main body 11 of the insect 1.
- a feature point is a characteristic point on the appearance of the insect 1, such as the boundary between parts, the base of the wings or the pattern of the insect 1, or the center position of the tip and rear end of each part of the insect 1. It is a position that can be determined based on a characteristic point within a part.
- the center position between the tip and rear end of each part of the insect 1 is, for example, the center position between the tip and rear end of the abdomen, or the center position between the tip and rear end of the body.
- the position of the intermediate point 43 is specified with the central part in the width direction of the boundary position between the thorax 15 and abdomen 16 as the feature point.
- the position of the intermediate point 43 is specified using the base of the wings of the body 14 as a feature point.
- the boundary between the parts of the body 11 of the insect 1 corresponding to the insect data D can be recognized, the boundary between the parts is treated as a feature point, and the insect data D is If the boundary between the parts of the body 11 of the corresponding insect 1 cannot be recognized, and a distinctive point within the part can be recognized, the distinctive point within the part is treated as a feature point.
- the order information adding step S4 is a step of adding order information 5 regarding the order to the three points of point information data 4 given in the point information adding step S3.
- the order information 5 is information regarding the order given to the point information data 4 in order to specify the posture of the insect 1 corresponding to the insect data D, and specifically, the head corresponding point 41, the intermediate point 43 , and information regarding the order in which the abdominal corresponding points 42 are arranged in a predetermined order.
- the order information 5 is information regarding the order in which the body main body 11 of the insect 1 is lined up in the overall length direction, and in this embodiment, as shown in FIGS. 6A and 6B, from the tip side (head side) to the rear end side ( This is information regarding the order of heading toward the ventral side.
- order information 5 is added to the point information data 4 based on the order of the parts of the body 11 of the insect 1, and in this embodiment, the head corresponding point 41, the intermediate point 43 , order information 5 regarding the order of the abdomen corresponding points 42 is provided.
- the order information data 5 is supplementary information given to the insect data D.
- order information 5 is added to the point information data 4, so insect data D to which point information data 4 is not added in the point information adding step S3 (part extraction Order information 5 is not assigned to the insect data D) for which it is determined that the part cannot be determined in S2.
- Image data for insect identification as described above is used to identify insects. Further, as described later, the image data for insect identification is used for machine learning. In any of the above-mentioned usage methods, the posture of the insect can be grasped by the point information data 4 and the order information data 5, so that the insect identification accuracy is improved.
- image data for insect identification is generated by performing each process in the above steps on unprocessed image data.
- the insect identification system 6 is a system that identifies the insect 1 included in the image data transmitted from the client terminal C (in this embodiment, the imaging device 3) via the Internet I.
- the insect identification system 6 is a system that processes target image data obtained by a client imaging an insect-captured sample, identifies and outputs the insect 1 included in the target image data.
- the target image data is captured by the above-mentioned imaging device 3 and transmitted to the insect identification system 6 using a communication means built into the imaging device 3.
- the insect identification system 6 includes a communication unit 61 that communicates with the outside, an image data acquisition unit 62 that acquires target image data from the outside, and identifies the type of insect 1 included in the target image data. It includes an insect identification section 63 and an output section 64 that outputs the identification result of the insect identification section 63.
- the image data acquisition unit 62 acquires target image data from the client terminal C (imaging device 3) via the communication unit 61 and the Internet I.
- the output unit 64 outputs the identification result to the client terminal C (specifically, the client's computer P, tablet terminal, etc.) via the communication unit 61 and the Internet I.
- such insect identification system 6 is realized by a computer.
- a computer includes a communication unit 61 that communicates with the outside, an image data acquisition unit 62 that acquires target image data from the outside, and a computer that identifies insects 1 included in the target image data.
- the insect identification unit 63 is configured to function as an insect identification unit 63 that identifies the type, and an output unit 64 that outputs the identification result of the insect identification unit 63.
- the insect identification unit 63 is configured to identify the type of insect 1 included in the target image using a learned model 65 constructed by machine learning using teacher data.
- the trained model 65 is constructed by machine learning using image data for insect identification generated by the above-described method for generating image data for insect identification as training data.
- the training data of the trained model 65 specifies at least three points for the unprocessed image data as the training image data in which the insect 1 is included in the image and the insect data D regarding the insect 1 included in the training image data. It includes posture information including given point information data 4 and order information 5 about the order given to the given point information data 4, and type information about the type of insect 1 included in the teacher image data.
- the point information data 4 is the same as the point information data 4 given in the point information adding step S3
- the order information 5 is the same as the order information 5 given in the order information adding step S4, and the type information is This is the same as the type information provided in the type information adding step.
- the teacher data of this embodiment is data to which frame information regarding the frame F surrounding the insect 1 included in the teacher image data is added.
- the frame information is the same as the frame information added in the frame information adding step.
- the trained model 65 is constructed by so-called supervised learning using the insect identification image data generated by the insect identification image data generation method described above as training data. That is, the trained model 65 is a model constructed by providing a large amount of insect identification image data with type information added to a machine learning algorithm.
- the insect identification unit 63 processes the target image data, discovers the insect 1 included in the target image data, and identifies the type of the discovered insect 1.
- the insect identification unit 63 of this embodiment adds type information of the discovered insects 1 to the target image data, and outputs the number of insects 1 for each type as an identification result. In this embodiment, the type of insect 1 is classified at the family level.
- point information data 4 and order information 5 are assigned to the extracted insect 1, so that the point information data 4 and order information 5 are Based on this, the posture of the insect 1 can be determined, and for example, even if the insect 1 is bent, the type of the insect 1 can be appropriately identified. Therefore, the identification accuracy of the insect 1 included in the image data can be improved.
- the point information data 4 is specified for two or more parts and the point information data 4 is given based on the order in which the parts are arranged, the posture of the insect 1 can be appropriately determined when the insect 1 is bent.
- the posture of the insect 1 can be determined based on at least three points including the head corresponding point 41, the abdomen corresponding point 42, and the intermediate point 43, the posture of the insect 1 can be appropriately determined even when the insect 1 is bent. .
- the point information data 4 is given to the feature point as the intermediate point 43, the point information data 4 can be given to substantially the same position on the insect 1 for insects 1 of the same type.
- the imaging device 3 since the imaging device 3 emits light from both sides of the travel area, it is possible to suppress the insect 1 from being in the shadow, and since it connects a plurality of images to generate unprocessed image data, the unprocessed image It is possible to prevent data bug 1 from becoming blurry.
- the image data including the bent insect 1 is used as training data. Can be used.
- the part extraction step S2 includes extracting a part of the insect 1 corresponding to the insect data D from the insect data D extracted before the point information adding step S3, and the part extraction step S2 includes at least the head of the insect 1.
- the portion 13 and abdomen 16 are extracted. Therefore, in the point information providing step S3, point information data 4 can be appropriately provided to the site of the body 11 of the insect 1.
- the part extraction step S2 also includes a determination step S22 for determining whether or not a part can be extracted from the extracted insect data D. Part extraction is performed only for the insect data D that is determined to be possible, and in the point information adding step S3, point information data 4 is given only to the insect data D from which the part has been extracted. Therefore, the point information data 4 can be reliably assigned to the part of the insect 1 that corresponds to the insect data D.
- the point information adding step S3 at least three points on the main body 11 of the insect 1 are designated and point information data 4 is added. Therefore, the posture of the insect 1 can be reliably determined regardless of the state of the appendage 12, such as the appendage of the insect.
- order information 5 is provided in the order of the head corresponding point 41, the intermediate point 43, and the abdomen corresponding point 42, or in the order of the abdomen corresponding point 42, the intermediate point 43, and the head corresponding point 41. Therefore, since the order information 5 is provided along the entire length direction of the body main body 11, the posture of the insect 1 can be determined reliably.
- the learned model 65 stores point information data 4 and order information 5 regarding at least three points specified for the insect data D included in the teacher image data. Since it is constructed by machine learning using teacher data including posture information, the type of insect 1 can be appropriately identified even if the insect 1 is bent, for example.
- the inventor has developed an insect identification system 6 using the above-mentioned trained model 65, which does not include information regarding the posture of the insect such as the point information data 4 and the order information 5, and which does not include information regarding the posture of the insect such as the point information data 4 and the order information 5.
- the results of a comparison between the system and a conventional insect identification system using a conventional trained model constructed as training data with only information added are described below.
- insect 1 was identified using the same captured insect sample, the identification accuracy of the conventional insect identification system was approximately 60.0%, whereas the identification accuracy of insect 1 using the trained model 65 according to the present invention was The identification accuracy of identification system 6 was approximately 79.1%.
- insect 1 when insect 1 is concentrated in the pre-processing image, insect 1 cannot be determined as insect 1, and the accuracy of detecting insect 1 decreases (specifically, 90.0% However, the insect identification system 6 using the trained model 65 maintains the accuracy of detecting insect 1 even when the insects 1 are crowded together. (approximately 90.0% or more).
- the method for generating image data for identifying this insect can also be used as pre-processing in the identification process.
- the method for generating image data for insect identification according to the present invention is applied to sample image data obtained by imaging the captured sample 2, and the sample image data as unprocessed image data is used during identification work. It can also be used as data for identification work as image data for insect identification.
- the machine can also be configured to provide the information. If the machine is configured to add information to the unprocessed image data in all steps, the computer will perform the insect extraction step S1, the part extraction step S2, the point information addition step S3, and the order information addition step. It can be configured as an insect identification image generation method that executes S4 to generate an insect identification image.
- each step is executed for all insects 1 before proceeding to the next step, this is not the only case; for example, all the steps may be executed for one insect 1. It can also be configured to perform all the steps for the next insect 1 after that. Each step can also be performed independently on a plurality of insects 1.
- a type information adding step is included in which type information regarding the type of insect 1 is added to the insect data D extracted in the insect extraction step S1
- the configuration may be such that the type information of the insect 1 is not provided.
- the insect extraction step S1 a case has been described in which frame information regarding a rectangular frame F surrounding the insect 1 in the image is provided as the frame information adding step, but the structure is not limited to such a configuration. It is also possible to add frame information regarding a frame other than a rectangular shape, such as a shape that follows the outline of insect 1, or it is also possible to not add frame F, that is, to not include the frame information adding step. .
- part extraction step S2 it is determined whether or not part extraction is possible for insect 1 corresponding to insect data D, and the part of insect 1 is extracted only for insect data D for which it has been determined that part extraction is possible.
- the configuration is not limited to this, and a configuration may be adopted in which the parts of the insect 1 are extracted for all insect data D.
- the part extraction step S2 is provided, a configuration may also be adopted in which the part extraction step S2 is not provided. If the part extraction step S2 is not provided, for example, point information data 4 is added to the insect data D in the point information provision step S3 at three points: one end, the other end, and a position between both ends of the body body 11. It is also possible to provide the point information data 4 with the order information 5 in order from one end toward the other end in the order information adding step S4. In this case, the point information data 4 is essentially provided so that at least three points correspond to two or more parts of the insect 1.
- one end of the body main body 11 can also be determined as a place where the antennae of the insect 1 are located.
- three or more point information data 4 are given to the body 14, and, for example, order information 5 is given sequentially from the head 13 side. You can also do that.
- the point information data 4 is provided only to the body main body 11, the point information data 4 is not limited to such a configuration, and the point information data 4 may also be provided to the antennae, legs, and wings as the appendages 12. can.
- one or more point information data 4 may be provided to each of the tip of the antennae, the main body 11, and the legs, and the order information 5 may be provided sequentially from the antennae.
- the posture of insect 1 can be determined and the identification accuracy of insect 1 can be improved. can be increased. Note that when four or more point information data 4 are provided, the posture can be determined with high accuracy.
- the insect data D can be provided with line information connecting the point information data 4 according to the order information 5 provided in the order information adding step S4.
- line information is provided, the attitude of the insect 1 can be easily determined based on the provided line information.
- information regarding the area surrounded by lines connecting the points given in the point information provision step S3 can also be given to the insect data D.
- information about the area is given, for example, if the area surrounded by the three point information data 4 given in the point information giving step S3 is large, it is determined that the insect 1 is largely bent. can do.
- the image data acquisition unit 62 in the insect identification system 6 has been described for acquiring image data via the Internet I, the configuration is not limited to this, and the image data acquisition unit 62 in the insect identification system 6 can acquire image data directly from a device that captures an image of insect paper or the like. It can also be configured to retrieve
- the configuration is not limited to this, and the image data can also be imaged using a method different from that of the teacher image data.
- the pest control system 7 is a system that performs treatment to suppress the harmful effects of insects when the number of insects has increased or there is a high possibility that the number of insects will increase. This is a system for controlling insects and/or insects that invade buildings.
- the pest control system 7 of this embodiment uses the insect identification system 6 described above.
- the pest control system 7 includes countermeasure control (feedback control) that performs processing to reduce insects inside the building and/or insects invading the building in response to an increase in insects; If there is a possibility that the number of insects will increase, preventive control (feedforward control) is carried out to prevent the increase in insects.
- countermeasure control feedback control
- preventive control feedforward control
- the pest control system 7 that performs countermeasure control includes a control unit 71 that receives and processes information regarding identification results from the insect identification system 6, and a storage unit 72 that stores various information.
- the processing result of the control section 71 is output to the processing section 8.
- the insect identification system 6 that outputs information regarding identification results to the pest control system 7 of the present embodiment is installed inside and outside the building, and is generated by an imaging device 3 provided for fixed point observation (specifically, and image processing), the number of each type of insect included in the target image data is output to the control unit 71 as an identification result.
- the imaging device 3 generates target image data by capturing images of insects caught by a plurality of insect catching devices (for example, insect catching paper) provided inside and outside the building. That is, the insect identification system 6 of this embodiment outputs identification results regarding the number of each type of insects caught by the insect trapping device.
- the insect identification system 6 is configured to output identification results at predetermined time intervals.
- the pest control system 7 includes a control unit 71 that executes predetermined processing, and a storage unit 72 that stores information regarding the identification results received from the insect identification system 6.
- the pest control system 7 of the present embodiment identifies the types of insects in which the number of insects has increased by more than a threshold value based on the information regarding the identification results received from the insect identification system 6, and uses a treatment method corresponding to the identified insect types. This is a system configured to output D4.
- the pest control system 7 of this embodiment is realized by cloud computing, it is not limited to this.
- the storage means 72 stores at least the previously received identification results received from the insect identification system 6. Furthermore, the storage means 72 can transmit the stored identification results of the previously received data to the control section 71 . Furthermore, the storage means 72 stores a database (for example, table T1 shown in FIG. 13) used by the control section 71 for various determinations, and transmits the database to the control section 71 in response to a request from the control section 71. As shown in FIG. 13, the database stored in the storage means 72 of this embodiment includes, for each type of insect (type information), intrusion route information D2, source information D3, and information regarding the treatment method D4. It is a related database.
- the control unit 71 compares the identification results received from the insect identification system 6 with the previously received identification results stored in the storage means 72, and determines whether the number of insects for each type of insect has increased by more than a threshold value.
- An increase determination unit 711 and a source determination unit 712 that determines, for the types of insects for which the increase determination unit 711 has determined that the increase has exceeded a threshold, invasion route information D2 regarding the invasion route of the insects of the relevant type and source information D3 regarding the generation source. and a processing method determining unit 713 that determines a processing method D4 for reducing insects in the building based on the determination by the source determining unit 712.
- the increase determination unit 711 receives the previously received identification results from the storage means 72, and compares the previously received identification results with the currently received identification results. , it is determined whether the number of insects has increased by more than a threshold value for each type of insect. That is, the increase determination unit 711 of the present embodiment determines the number of insects newly captured by the insect trapping device from the time when the insect identification system 6 outputs the previous identification result until the current identification result is output. It is determined for each type of insect whether or not it is equal to or greater than a threshold value.
- the number of insects newly captured by the insect trapping device is equal to or greater than the threshold value in a predetermined time (interval at which the insect identification system 6 outputs identification results). Since it can be determined for each type of insect, if there are many insects of a particular type in the environment where the insect trap is installed, it can be determined that there are many insects of that type.
- the threshold value in this embodiment is a preset numerical value, the configuration is not limited to this, and for example, a numerical value twice the amount of increase during normal times may be used as the threshold value.
- the source determining unit 712 determines the source information D3 regarding the source of the insects of the type. That is, when there are many insects of a specific type in the environment where the insect trap is installed, the source determining unit 712 identifies the source of the insects of the particular type.
- the source determining unit 712 of the present embodiment determines intrusion route information D2 regarding the intrusion route of the type of insect that has been determined to have increased, and source information D3 regarding the source of the insect.
- the invasion route information D2 is information that distinguishes whether the type of insect is an external invasion type that occurs outside the building and invades the building, or an internal insect type that occurs inside the building.
- the source determining unit 712 of the present embodiment obtains the increased types of insect invasion route information D2 and generation source information D3 based on a database prepared in advance regarding the invasion route information D2 and the generation source information D3 for each type of insect. Identify.
- the source information D3 is information for identifying the source of the insect of that type within the building. This is information for specifying the method, and for internally generated insects, it is information for specifying the environment inside the building where they occur. Specifically, the source information D3 includes insects that fly and invade buildings, and include flying large flies that belong to large flies (for example, house flies, black flies, black flies, etc.).
- flying insects that enter buildings by flying and other than large flies e.g., insects of the order Hyperformes other than large flies, thrips, moths, flycatchers, fly ants, bees
- insects that invade buildings by wandering or creeping e.g., ants, earwigs, crickets, springtails, centipedes, millipedes, etc.
- the source information D3 includes insects that are generated from water systems such as drainage systems (e.g., butterflies, fruit flies, flea flies, false flies, and hayato flies). , black flies, etc.), mycophagous insects that originate from fungi (e.g., winged chatatates, cylindrical beetles, serpent beetles, snail beetles, etc.), and insects that originate from indoor dust and food.
- water systems e.g., drainage systems (e.g., butterflies, fruit flies, flea flies, false flies, and hayato flies).
- mycophagous insects that originate from fungi e.g., winged chatatates, cylindrical beetles, serpent beetles, snail beetles, etc.
- insects that originate from indoor dust and food e.g., winged chatatates, cylindrical beetles, serpent beetles, snail beetles, etc
- Indoor dust and food systems including, for example, chatate beetles, snails, cutworms, white mites, grasshoppers, flatullies, leopard beetles, and caterpillars
- insect predators that feed on insects
- insects of the order Araneidae such as Arachnidae, Araneidae, and Piperidae belong to this category
- heat source/interstitial insects that generate from heat sources or gaps
- cockroaches, etc. belong to this category
- the source determining unit 712 of this embodiment refers to the table T1 as shown in FIG. 13, and determines the intrusion route information D2 and the source information D3 for the type that the increase determining unit 711 determines has increased. Specifically, with reference to the type information D1 of the table T1, the intrusion route information D2 and source information D3 corresponding to the type determined to have increased are acquired. Specifically, if the type of insects determined to have increased is one of a1, a2, or a3, the source determination unit 712 determines that the insects that are increasing are externally invasive types and the source is flying. It is determined that it is a type of large fly.
- the source determining unit 712 outputs only one type as the type of insects that are increasing. For example, if the types of insects that have been determined to have increased are a4, a5, a6, a7, and a8, the source determination unit 712 determines that the insects that are increasing are external invasive types and that the source of the insects is is determined to be of the flying type or other type.
- the treatment method determining unit 713 determines and outputs a treatment method D4 for reducing insects based on the source information D3 determined by the source determining unit 712.
- the treatment method determination unit 713 of this embodiment outputs both the treatment method D4 using physical control D41 and the treatment method D4 using chemical control D42 (chemical treatment) as the treatment method D4 for reducing insects.
- the processing method determination unit 713 of this embodiment determines the processing method D4 by referring to the database regarding the processing method D4, and specifically, the processing method D4 is determined by referring to the table T1 as shown in FIG. judge.
- the processing method determining unit 713 uses a processing method corresponding to the other types. As physical control D41 in D4, confirmation of a passage that can become an external entry hole is output. In this manner, the processing method determination unit 713 outputs the processing method D4 for each type in the source information D3, dividing it into physical control D41 and chemical control D42.
- the treatment method determining unit 713 determines the type and amount of the chemical to be used, as well as the chemical treatment method and location, and determines whether or not to immediately perform the chemical treatment using the determined method. to judge. The specific process flow will be described later.
- the processing unit 8 is configured to perform predetermined processing according to the output of the processing method determination unit 713.
- the processing unit 8 of this embodiment is a notification unit configured to notify the manager of the building to be controlled, etc. of the processing method D4 regarding the physical control D41 and the chemical control D42 output by the processing method determination unit 713.
- 81, and a chemical processing means 82 that executes chemical processing related to the chemical control D42 outputted by the processing method determination unit 713.
- the pest control system 7 includes a current result acquisition step S51, a previous result acquisition step S52, an increase determination step S53, a source determination step S54, a treatment method determination step S55, and a treatment output step S56. and repeat.
- the current result acquisition step S51 is a step of acquiring the identification results output by the insect identification system 6 and acquiring the number of insects of each type included in the identification results.
- the current result acquisition step S51 of the present embodiment is executed by the increase determination unit 711, and the insect identification system 6 outputs the identification result every predetermined time, so that the identification result can be obtained every predetermined time.
- the acquired identification result is stored in the storage means 72.
- the previous result acquisition step S52 is a step of acquiring the identification results outputted last time by the insect identification system 6, and acquiring the number of insects for each type in the previous identification results included in the identification results.
- the previous result acquisition step S52 of the present embodiment is executed by the increase determination unit 711 to acquire the previous identification result from the storage means 72.
- the increase determination step S53 the number of insects of each type included in the identification result outputted this time by the insect identification system 6 is compared with the number of insects of each type included in the identification result outputted last time, and a threshold value is set for each type. This is a step of determining whether the number of insects has increased by a threshold value or more, and outputting the type of insects for which the number of insects has increased by a threshold value or more. If it is determined in the increase determination step S53 that there is no type for which the number of insects has increased by more than the threshold value (NO in the increase determination step S53), the process moves to the current result acquisition step S51, and the number of insects has increased by more than the threshold value. If it is determined that there is one or more types (YES in the increase determination step S53), the process moves to the source determination step S54. Further, the increase determination step S53 of this embodiment is executed by the increase determination unit 711.
- the source determination step S54 is a step of determining the intrusion route information D2 and the source information D3 for the insects for which it was determined in the increase determination step S53 that the number of insects has increased by more than the threshold value.
- the database (table T1 shown in FIG. 13 in this embodiment) is referred to for the types for which it was determined that the number of insects has increased by more than the threshold value in the increase determination step S53.
- This is a step of referring to the intrusion route information D2 and the source information D3 to identify the types of sources that are increasing in number.
- the increase determination step S53 if one or more types of a1, a2, and a3 are increasing, it is determined that the type of externally invading large flying flies is increasing, and a4, If one or more types of a5, a6, a7, and a8 are increasing, it is determined that external invasion type flying and other types are increasing, and b1, b2, b3, b4, b5 If one or more of these types is increasing, it is determined that the wandering/crawling type of external invasive type is increasing, and if c1 or c2 is increasing, it is determined that the external invasive type is increasing. It is determined that other types of cases are increasing.
- the generation source determination step S54 if one or more types of d1, d2, d3, d4, and d5 are increasing, the types of internally generated drainage/plumbing systems are increasing. If one or more types of e1, e2, e3, e4, and e5 are increasing, it is determined that the endogenous bacteriophagous type is increasing, and f1, f2 , f3, f4, and f5, it is determined that the type of internally generated indoor dust/food is increasing, and one of g1, g2, and g3 is increasing.
- the generation source determination step S54 the generation source is determined for all types for which it was determined in the increase determination step S53 that the amount has increased by a threshold value or more. Further, the source determining step S54 is executed by the source determining unit 712.
- the treatment method determination step S55 is a step of determining the treatment method D4 that corresponds to the source of the increasing number of types of insects. Specifically, this is a step of determining a process for reducing the increasing number of insects based on the intrusion route information D2 and the source information D3 determined in the source determining step S54.
- the processing method D4 corresponding to the source information D3 is output with reference to a database (for example, table T1 shown in FIG. 13). Further, in the treatment method determination step S55, the physical control D41 and the chemical control D42 in the treatment method D4 are output separately.
- physical control D41 includes checking for dead animals that could be the source, installing insect nets, and checking openings. , and outputs caution instructions such as the shutter opening time, and outputs instructions for spraying the chemical into the space at the entry point and/or dispersing the chemical at the source as chemical control D42. Output. If flying insects or other types of pests are increasing, physical pest control D41 should be carried out by installing insect nets, checking openings, giving instructions on how long to open shutters, etc., and instructing the placement of attraction lights and LED switching lights around buildings.
- chemical control D42 an instruction for spraying the chemical into the space at the entry point and/or dispersing the chemical at the source is output.
- physical control D41 is the installation of sticky traps
- chemical control D42 is spraying of chemicals around the building and preventive chemical treatment of areas where they can invade. If the number of other types of external intrusion type is increasing, we support confirmation of passages that can become external intrusion holes as physical control D41.
- a cleaning instruction for the position corresponding to the source information D3 is output as physical pest control D41, and a chemical As target control D42, chemical treatment for exterminating the insects and preventive chemical treatment for the source of the outbreak are output.
- physical control D41 if the types of drainage and water systems are increasing, removal of the source such as drainage and water systems is performed, and the number of fungivorous systems is increasing. If there is an increase in indoor dust and food-based dust, cleaning and removing mold, etc. that could be the source, and cleaning and removing dust that could be the source of insect-predatory dust, etc.
- a notification of bed bug detection will be sent as physical control D41, and a notification of bed bug detection will be sent as chemical control D42 to exterminate the infested insects.
- the treatment method determination step S55 if the number of external invasive types is increasing, excluding other types, a method of preventing the intrusion of the increasing number of insects is selected as physical control D41. Then, as chemical control D42, chemical treatment for at least one of the invaded part and the source is output. In addition, if the number of internally occurring types is increasing, excluding other types, cleaning and removal of the source is output as physical control D41, and chemical treatment to exterminate the insects as chemical control D42. and output preventive drug treatment for the source.
- the treatment method determination step S55 regarding the chemical treatment as chemical control D42, the type and amount of the chemical to be used, the method of chemical treatment, and the location where the treatment is performed are determined. Specifically, as shown in FIG. 11, in the drug treatment, there is a treatment method determining step S57 in which it is determined which type of drug to use and in what method to perform the treatment, and a step S57 in which the amount of the drug used in the treatment is determined. A processing amount determination step S58 is performed, and an implementation feasibility determination step S59 is performed to determine whether or not to immediately perform drug treatment.
- the treatment method determination step S57 it is determined to spray the chemical selected according to the insect source to the insect source and the location where the insect was detected. For example, a drug processing method corresponding to the source information D3 is determined based on a database (not shown).
- the amount of the drug selected in the treatment method determination step S57 to be used for the treatment is determined.
- the amount of chemical used for treatment can be determined depending on the method of chemical treatment and the number of insects generated. For example, it can be configured to derive and determine the treatment method D4 and the amount of chemical treatment corresponding to the number of insects based on a database (not shown).
- the implementation feasibility determination step S59 is a step of determining whether or not to execute the drug treatment. Specifically, in the implementation feasibility determination step S59, it is determined whether or not the drug treatment is to be performed, depending on the situation of the location where the drug treatment is to be performed. In the implementation feasibility determination step S59 of this embodiment, as shown in FIG. A time determination step S592 for determining whether time has elapsed, and a processable date and time determination step S593 for determining whether it is the date and time when spraying of the medicine is permitted are executed. In the person determination step S591, for example, information from a human sensor or camera installed at the location where the drug treatment is to be performed is obtained, and it is determined whether or not there is a person at the location where the drug treatment is to be performed.
- the processing possible date and time determination step S593 is a step of determining whether or not it is a preset date and time when medicine treatment is possible.
- the medicine processing possible date and time is set based on the time or the day of the week.
- the processing unit 8 is instructed to execute drug processing (drug processing execution instruction step S594). Further, in the person determination step S591, it is determined that there is a person (NO in the person determination step S591), or in the time determination step S592, it is determined that a predetermined time has not elapsed since the previous drug treatment (time determination step S592).
- step S593 if it is determined in the processing possible date and time determination step S593 that the processing is not possible (NO in the processing possible date and time determination step S593), it is determined that the drug treatment is not to be performed immediately, and the skip processing is performed in step S595. Execute. In the skip process S595, the notification means 81 of the processing unit 8 notifies the administrator etc. of the necessity of drug processing, and allows the administrator to manually execute the drug processing or to specify the date and time for the drug processing. Encourage them to specify.
- various instructions are given to the processing unit 8 based on the content output in the processing method determination step S55.
- the notification means 81 notifies the building manager etc. of the output content, and the manager etc. performs the physical pest control D41. Encourage them to do it.
- the chemical processing means 82 is controlled to execute chemical processing or skip processing S595 is executed.
- the notification is made by the notification means 81, it is also possible to notify that the notification is for countermeasure control.
- the pest control system 7 of the present embodiment based on the identification results output by the insect identification system 6, it is determined whether there is a type of insects whose number has increased by more than a threshold value, and If there is an increased number of types, the processing method D4 corresponding to the increased number of types is output, so that the harmful effects of insects can be suppressed.
- the source information D3 for the increased types is determined and the processing method D4 corresponding to the source information D3 is output, so for example, when multiple types of insects are generated from the same source, It is possible to suppress outputting a duplicate processing method D4.
- the present invention is not limited to such a configuration.
- the configuration is not limited to this configuration. It can also be configured to use the identification results determined by. Insect identification is not limited to using the identification results of insects captured by insect traps, but also detects insects moving within a building, identifies the type of the insect, and outputs the identification results for each type.
- a system 6 for example, an insect identification system 6 that outputs the type and number of insects as identification results based on an odor sensor) can also be employed.
- the pest control system 7 has been described for the case where it determines and outputs the processing method D4, it is not limited to such a configuration.
- it outputs the intrusion route information D2 and the source information D3, and the control system It can also be configured to notify the following.
- a building manager or the like can perform the necessary work for pest control based on the notified intrusion route information D2 and source information D3.
- the configuration is not limited to this, for example.
- the processing method determining unit 713 may be configured to output the processing method D4 corresponding to the type whose number has been increased, which is output by the increase determining unit 711, without being based on the source information D3.
- the processing method determination unit 713 outputs both the physical pest control D41 and the chemical pest control D42 as the pest control method
- the processing method determination unit 713 outputs only one of the physical pest control D41 and the chemical pest control D42. It can also be configured as When adopting such a configuration, the treatment method determination unit 713 selects and outputs the appropriate treatment method D4 from the physical control method D41 and the chemical control method D42 based on information such as the number of insects. It can also be configured.
- processing method D4 of the physical pest control D41 has been described for the case where the building manager etc. is notified by the notification means 81, the method D4 is not limited to such a configuration. It can also be configured.
- the processing unit 8 may be configured to close the opening when confirmation of the opening is output.
- the increase determination step S53 a case has been described in which the determination is made using the previous identification results, but the configuration is not limited to this.
- the number of insects in the past several identification results is It is also possible to determine whether the current number has increased by more than a threshold value with respect to the average value.
- the processing method D4 is determined based on the source information D3, the configuration is not limited to this, and for example, it may be configured to be determined based on the type information D1. It can also be configured to be determined based on the intrusion route information D2.
- the pest control system 7 that performs preventive control uses the identification results received from the insect identification system 6 and the environmental information regarding the environment inside and outside the building received from the factor sensor unit 9 that monitors the environment inside and outside the building.
- a control unit 71 that determines whether there is a risk of an increase in insects based on the information, and outputs a preventive method for preventing an increase in insects when there is a risk of an increase in insects; and various determinations by the control unit 71. 13 and 16) and a storage means 72 for storing environmental information received from the factor sensor.
- the configuration of the insect identification system 6 and the configuration of the processing unit 8 are similar to the configuration of the pest control system 7 that executes countermeasure control.
- the factor sensor unit 9 is a sensor that monitors the environment inside and outside the building, and specifically, it is a sensor that monitors at least humidity and temperature.
- the factor sensor unit 9 of this embodiment detects an environment in a building where internally occurring insects can occur (specifically, a location indicated by source information D3 for internally occurring insects shown in FIG. 13). ) and the environment outside the building (specifically, around the building). Further, the factor sensor unit 9 outputs environmental information regarding the environment obtained through monitoring to the pest control system 7.
- the storage means 72 stores the environmental information received from the factor sensor unit 9 for a predetermined period (for example, one year).
- the storage means 72 also stores a database (for example, tables T1 and T2 shown in FIGS. 13 and 16) used by the control section 71 for various determinations, and transmits the database to the control section 71 in response to a request from the control section 71.
- the storage means 72 stores a database in which insect types and increase factors (prevention conditions D7) are associated with each other, as shown in FIG. 16, and a database as shown in FIG. 13 (type information D1). , a database in which intrusion route information D2, source information D3, and information regarding processing method D4 are associated is stored.
- the growth zero point D5 and the effective cumulative temperature D6 necessary for generation are associated with the insect type.
- the control unit 71 includes an insect species extraction unit 714 that extracts types of insects that exist in a number equal to or greater than a threshold value from the identification results received from the insect identification system 6, and a factor that increases the number of existing types extracted by the insect species extraction unit 714.
- a factor extraction unit 715 that extracts (preventive condition D7), a condition determining unit 716 that determines whether the increasing factor (preventive condition D7) extracted by the factor extracting unit 715 is satisfied, and a condition determining unit 716
- a prevention method determining unit 717 is provided, which determines a method for preventing an increase in types that are determined to satisfy an increase factor (prevention condition D7) and outputs the result to the processing unit 8.
- the insect species extraction unit 714 refers to the number of each type of insect output as an identification result from the insect identification system 6, and extracts types whose number is equal to or greater than a threshold value.
- the insect species extraction unit 714 of this embodiment is configured to extract, from the identification results, species in which one or more insects exist.
- the factor extraction unit 715 extracts a factor (prevention condition D7) that increases the number of types extracted by the insect species extraction unit 714.
- the factor extraction unit 715 refers to the database stored in the storage unit 72 (specifically, table T2 shown in FIG. 16), and determines the factors ( Preventive condition D7) is extracted.
- the factor extraction unit 715 extracts as a factor the condition recorded as the prevention condition D7 in the table T2 shown in FIG.
- One or more of the following conditions are extracted as factors: numerical value (in this embodiment, day degree), average temperature (in this embodiment, daily average temperature), and relative humidity.
- the factor extraction unit 715 sets the fact that the effective cumulative temperature is higher than 250 degrees Celsius for type a1 as an increasing factor (preventive condition D7), and sets the effective cumulative temperature higher than 1300 degrees Celsius for type a4 as an increasing factor (prevention condition D7).
- the effective cumulative temperature is higher than 300 days and the daily average temperature is lower than 20 degrees as an increasing factor (prevention condition D7).
- condition D7) for type f1, the effective cumulative temperature is higher than 50 days and the relative humidity is less than 90% are extracted as increasing factors (prevention condition D7).
- the factor extraction unit 715 extracts information (specifically, the growth zero point D5) for determining whether each of the above factors is satisfied from the database.
- the condition determination unit 716 determines for each type whether a factor that increases the number of insects (prevention condition D7) is satisfied based on the environmental information output by the factor sensor unit 9 and the factors extracted by the factor extraction unit 715. do. Specifically, the condition determination unit 716 uses the environmental information output by the factor sensor unit 9 and stored in the storage unit 72, and information for determining whether each factor extracted by the factor extraction unit 715 is satisfied. (for example, growth zero point D5), the average temperature, relative humidity, effective integrated temperature, etc. are derived, and based on the derived results, it is determined whether the factors extracted by the factor extraction unit 715 are satisfied.
- condition determination unit 716 determines whether the factors are satisfied based on environmental information outside the building for insects that invade the outside, and determines whether the factors are satisfied for insects that occur inside the building. It is determined whether the factors are satisfied based on environmental information (specifically, environmental information about places where internally generated insects can occur in the building).
- the prevention method determining unit 717 determines a method for preventing an increase in the types for which the condition determining unit 716 determines that the factors are satisfied.
- the preventive method determining unit 717 refers to a table T1 as shown in FIG. 13 and determines a preventive method for preventing an increase in the number of insects for the types for which the condition determining unit 716 determines that the factors are satisfied.
- the prevention method determining unit 717 of this embodiment determines the intrusion route information D2 and source information D3 corresponding to the type determined to satisfy the factor, and specifically refers to the type information D1 of the table T1. Then, intrusion route information D2 and source information D3 corresponding to the type determined to have increased are acquired. Furthermore, the preventive method determining unit 717 determines a preventive method for preventing an increase in insects based on the acquired source information D3, and outputs the determined preventive method to the processing unit 8.
- the prevention method determination unit 717 of this embodiment outputs a treatment method D4, which is similar to the treatment method D4 output by the treatment method determination unit 713 described above, as a prevention method.
- the present invention is not limited to such a configuration, and may be configured to output different methods for the treatment method D4 and the prevention method.
- the prevention method it is possible to output only the physical control D41 without outputting the chemical control D42, or to output the adjustment of temperature and humidity (environment), or to output the adjustment of the temperature and humidity (environment).
- the pest control system 7 includes an identification result acquisition step S61, a type information extraction step S62, an increase factor extraction step S63, a factor information acquisition step S64, a factor determination step S65, and a source information acquisition step S63.
- Step S66, preventive method acquisition step S67, and process output step S56 are repeatedly executed.
- the identification result acquisition step S61 is a step of acquiring the identification results output by the insect identification system 6.
- the identification result acquisition step S61 of this embodiment is executed at predetermined time intervals. Moreover, the identification result acquisition step S61 of this embodiment is executed by the insect identification system 6 outputting the identification result at every predetermined time.
- the identification result acquisition step S61 is executed by the insect species extraction unit 714.
- the type information extraction step S62 is a step of extracting types whose number is equal to or greater than a threshold value from the identification results acquired by the identification result acquisition step S61. In this embodiment, this is a step of extracting the types of insects for which one or more insects are present in the identification results.
- the type information extraction step S62 is executed by the insect type extraction unit 714.
- the increasing factor extraction step S63 is a step of extracting an increasing factor (prevention condition D7) from the database for the type extracted in the type information extraction step S62.
- the increase factor extraction step S63 the database stored in the storage means 72 is referred to, and the factors that cause the number of insects of the type extracted in the type information extraction step S62 to increase (prevention condition D7) and the factors that cause the increase (prevention Information (specifically, growth zero point D5) for determining whether condition D7) is satisfied is extracted.
- the increase factor extraction step S63 is executed by the factor extraction unit 715.
- the factor information acquisition step S64 is a step of acquiring environmental information regarding the factor that increases the number of insects (prevention condition D7) extracted in the increase factor extraction step S63. Specifically, in the factor information acquisition step S64, the environmental information stored in the storage means 72 is acquired, and whether the acquired environmental information and the increasing factor (preventive condition D7) extracted in the increasing factor extracting step S63 are satisfied. This is a process of deriving factor information such as average temperature, relative humidity, and effective cumulative temperature based on information for determining the temperature. The factor information acquisition step S64 is executed by the factor extraction unit 715.
- the factor determination step S65 is a step of determining whether the factor information derived in the factor information acquisition step S64 satisfies the factor for increasing the number of insects (prevention condition D7) extracted in the increase factor extraction step S63. For all types extracted in the type information extraction step S62, if it is determined in the factor determination step S65 that the factor information does not satisfy the prevention condition D7 (NO in the factor determination step S65), the identification result acquisition step S61 is performed. If it is determined in the factor determination step S65 that the factor information satisfies the prevention condition D7 for at least one type extracted in the type information extraction step S62 (YES in the factor determination step S65), the source The process moves to information acquisition step S66. The factor determination step S65 is executed by the condition determination unit 716.
- the source information acquisition step S66 is a step of specifying the source information D3 and the invasion route information D2 for the types that satisfy the factor that increases the number of insects (prevention condition D7). Specifically, in the source information acquisition step S66, the database (table T1 shown in FIG. 13 in this embodiment) is referred to for the types of insects that are determined to satisfy the conditions for increasing the number of insects (prevention conditions D7). This is a step of identifying increasing types of sources by referring to route information D2 and source information D3. The specific process in the source information acquisition step S66 is the same as the source determining step S54 in the pest control system 7 that executes countermeasure control. The source information acquisition step S66 is executed by the prevention method determining unit 717.
- the preventive method acquisition step S67 is a step of determining a treatment method D4 corresponding to the source of the type of insect that is determined to satisfy the increasing factor. Specifically, this is a step of determining a process for suppressing an increase in the number of insects, based on the intrusion route information D2 and the source information D3 determined in the source information acquisition step S66.
- the processing method D4 corresponding to the source information D3 is output with reference to a database (for example, table T1 shown in FIG. 13).
- the prevention method acquisition step S67 of this embodiment is similar to the treatment method determination step S55 in the pest control system 7 that executes countermeasure control. That is, the prevention method acquisition step S67 includes a step of determining the specific method of chemical treatment in chemical control D42 and whether or not to implement the chemical treatment.
- the processing output step S68 provides various instructions to the processing unit 8 based on the content output in the preventive method acquisition step S67.
- the processing output step S68 of this embodiment is similar to the processing output step S56 in the pest control system 7 that executes countermeasure control. Specifically, when the physical pest control D41 is output in the prevention method acquisition step S67, the notification means 81 notifies the building administrator of the output content, and the manager etc. performs the physical pest control D41. Encourage them to do it. Further, when an output related to chemical control D42 is output, the chemical processing means 82 is controlled to execute chemical processing or skip processing S595 is executed.
- the pest control system 7 configured as described above, for a type of insects in which the number of insects is equal to or higher than a threshold value, it is determined whether or not the factors for increasing the number of insects of that type are satisfied, and if the factors for increasing the number of insects of that type are satisfied. Since insects can be treated for prevention, the increase in insects can be suppressed.
- the source information D3 is determined for the types that satisfy the factors that increase the number of insects, and the processing method D4 corresponding to the source information D3 is output. , it is possible to suppress outputting the duplicate processing method D4.
- the structure is not limited to this, and for example, information regarding the season, weather, and sunshine hours can be used. You can also.
- the pest control system 7 that performs countermeasure control and the pest control system 7 that performs preventive control have been described separately, the present invention is not limited to this configuration, and a single system may be configured to perform both countermeasure control and preventive control. You can also do it. In the case of such a configuration, for example, when the insect identification system 6 outputs an identification result, both countermeasure control and preventive control are executed, and the processing necessary for countermeasure and the processing necessary for prevention are performed. Both can be output to the processing section 8.
- This disclosure includes the following content.
- the insect extraction step includes a frame information adding step of adding frame information regarding a frame surrounding the insect included in the image, and the insect data is added based on the frame information added in the frame information adding step.
- the point information adding step is a step of adding point information data such that the at least three points correspond to two or more parts out of a plurality of parts that the insect has, and the order information adding step The method for generating image data for insect identification according to (1) or (2), wherein the order information is added to the point information based on the order in which the two or more parts are arranged.
- the point information adding step includes a head corresponding point corresponding to the head of the insect corresponding to the insect data, an abdomen corresponding point corresponding to the abdomen of the insect, and a head corresponding point and the abdomen corresponding
- the image data for insect identification according to any one of (1) to (3), wherein at least three points, intermediate points located between the points, are designated for the insect data and point information data is added to the insect data. Generation method.
- the point information adding step adds point information data to a feature point located between the head corresponding point and the abdomen corresponding point as the intermediate point, any one of (1) to (4). 1.
- the pre-processing image data is arranged at a distance from the capture surface so as to image the capture surface of a captured sample on which insects have been captured, and along the surface direction of the capture surface.
- a plurality of images are generated by an imaging device including a traveling camera and an irradiation device that irradiates light onto the capturing surface from both sides of a traveling area where the camera travels, and the imaging device runs the camera to generate a plurality of images.
- the method for generating image data for insect identification according to any one of (1) to (5), wherein the image data for insect identification is generated by capturing a plurality of images and connecting the plurality of captured images to generate the unprocessed image data.
- a method for generating image data for insect identification which generates image data for insect identification as training data used in machine learning, wherein type information regarding the type of insect is added to the extracted insect data.
- the method for generating image data for insect identification according to any one of (1) to (6), including the step of providing type information.
- An image data acquisition unit that acquires target image data in which an insect to be identified is included in the image; and an insect identification unit that identifies the type of insect included in the target image data, the insect identification unit is configured to identify the type of insect included in the target image data acquired by the image data acquisition unit using a learned model constructed by machine learning using teacher data, and the learned model Teacher image data that includes insects, point information data that specifies at least three points among the insect data related to insects included in the teacher image data, and point information data that is added to the insect data, and point information data that is added to the point information data.
- An insect identification system constructed by machine learning using training data including posture information including order information regarding the order in which the insects were placed, and type information regarding the types of insects included in the training image data.
- the method for generating image data for insect identification (1) since point information data 4 and order information 5 are given to the extracted insect 1, based on the point information data 4 and order information 5, The posture of the insect 1 can be determined, and even if the insect 1 is bent, for example, the type of the insect 1 can be appropriately identified. Therefore, the identification accuracy of the insect 1 included in the image data can be improved.
- point information data 4 is specified for two or more parts, and point information data 4 is given based on the order of the parts, so that insect 1 is bent.
- the posture of insect 1 can be appropriately determined when
- the posture of the insect 1 can be determined based on at least three points including the head corresponding point 41, the abdomen corresponding point 42, and the intermediate point 43. Even if the insect is bent, the posture of the insect 1 can be determined appropriately.
- point information data 4 is assigned to the feature point as the intermediate point 43, so for insects 1 of the same type, points are placed at approximately the same position on insect 1. Information data 4 can be added.
- the imaging device 3 emits light from both sides of the running area, so it is possible to suppress the insect 1 from being in the shadow, and to capture multiple images. Since the unprocessed image data is generated by connecting the images, it is possible to suppress blurring of the insect 1 in the unprocessed image data.
- image data for insect identification (7) since it includes a type information adding step of adding type information regarding the type of insect 1, even if the insect 1 included in the image data is bent, , image data including the bent insect 1 can be used as teacher data.
- the learned model 65 generates posture information including point information data 4 and order information 5 regarding at least three points specified for the insect data D included in the teacher image data. Since it is constructed by machine learning using the included training data, the type of insect 1 can be appropriately identified even if insect 1 is bent, for example.
- Order information 6...Insect identification system, 61...Communication unit, 62...Image data acquisition unit, 63...Insect identification unit, 64...Output unit, 65...Learned model, 7...Pest control system, 71...Control unit, 711...Increase determination 712... Source determination unit, 713... Treatment method determination unit, 714... Insect species extraction unit, 715... Factor extraction unit, 716... Condition determination unit, 717... Prevention method determination unit, 72... Storage means, 8... Processing Part, 81... Notification means, 82... Drug processing means, C... Client terminal, D... Insect data, F... Frame, I... Internet, P... Computer
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Pest Control & Pesticides (AREA)
- Artificial Intelligence (AREA)
- Medical Informatics (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Evolutionary Computation (AREA)
- Computing Systems (AREA)
- Health & Medical Sciences (AREA)
- Environmental Sciences (AREA)
- Insects & Arthropods (AREA)
- Wood Science & Technology (AREA)
- Zoology (AREA)
- Human Computer Interaction (AREA)
- Catching Or Destruction (AREA)
- Image Analysis (AREA)
- Debugging And Monitoring (AREA)
- Investigating Or Analysing Biological Materials (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
- Editing Of Facsimile Originals (AREA)
Abstract
Description
Claims (8)
- 画像中に虫が含まれる処理前画像データから一匹の虫に関する虫データを抽出する虫抽出工程と、
前記抽出された虫データの中で少なくとも3点を指定して点情報データを虫データに付与する点情報付与工程と、
該点情報付与工程によって指定された前記少なくとも3点の点情報データに順番に関する順情報を付与する順情報付与工程と、を有する虫同定用画像データの生成方法。 - 前記虫抽出工程は、前記画像中に含まれる虫を囲う枠に関する枠情報を付与する枠情報付与工程を含み、該枠情報付与工程で付与された枠情報に基づいて、虫データを抽出する請求項1に記載の虫同定用画像データの生成方法。
- 前記点情報付与工程は、虫が有する複数の部位のうち、2以上の部位に対して前記少なくとも3点が対応するように点情報データを付与する工程であり、
前記順情報付与工程は、前記2以上の部位の並び順に基づいて前記順情報を前記点情報に付与する請求項1に記載の虫同定用画像データの生成方法。 - 前記点情報付与工程は、前記虫データに対応する虫の頭部に対応した頭部対応点、前記虫の腹部に対応した腹部対応点、並びに、前記頭部対応点及び前記腹部対応点の間に位置する中間点、の少なくとも3点を前記虫データに対して指定して点情報データを付与する請求項1に記載の虫同定用画像データの生成方法。
- 前記点情報付与工程は、前記中間点として、前記頭部対応点及び前記腹部対応点の間に位置する特徴点に点情報データを付与する請求項4に記載の虫同定用画像データの生成方法。
- 前記処理前画像データは、捕獲面に虫が捕獲された捕獲済試料における前記捕獲面を撮像するように該捕獲面から離間して配置され、且つ前記捕獲面の面方向に沿って走行するカメラと、該カメラが走行する走行エリアの両サイドから前記捕獲面に光を照射する照射装置と、を備える撮像装置によって生成され、
該撮像装置は、前記カメラを走行させて複数の画像を撮像し、該撮像された複数の画像をつなげて前記処理前画像データを生成する請求項1に記載の虫同定用画像データの生成方法。 - 機械学習に用いられる教師データとしての虫同定用画像データを生成する虫同定用画像データの生成方法であって、前記抽出された虫データに対して、虫の種類に関する種類情報を付与する種類情報付与工程を含む請求項1に記載の虫同定用画像データの生成方法。
- 画像中に同定対象の虫が含まれる対象画像データを取得する画像データ取得部と、前記対象画像データに含まれる虫の種類を同定する虫同定部と、を備え、
前記虫同定部は、教師データを用いた機械学習によって構築された学習済モデルによって前記画像データ取得部が取得した前記対象画像データに含まれる虫の種類を同定するように構成され、
前記学習済モデルは、
画像中に虫が含まれる教師画像データと、
該教師画像データに含まれる虫に関する虫データの中で少なくとも3点を指定して虫データに付与された点情報データ、及び、該点情報データに付与された順番に関する順情報、を含む姿勢情報と、
前記教師画像データに含まれる虫の種類に関する種類情報と、を含む教師データを用いた機械学習によって構築される虫同定システム。
Priority Applications (5)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| KR1020247032942A KR20240170915A (ko) | 2022-03-31 | 2023-03-24 | 벌레 동정용 화상 데이터의 생성 방법 및 벌레 동정 시스템 |
| EP23780106.3A EP4483713A4 (en) | 2022-03-31 | 2023-03-24 | METHOD FOR GENERATING INSECT IDENTIFICATION IMAGE DATA AND INSECT IDENTIFICATION SYSTEM |
| AU2023242975A AU2023242975A1 (en) | 2022-03-31 | 2023-03-24 | Method for generating bug identification image data, and bug identification system |
| US18/849,274 US20260024371A1 (en) | 2022-03-31 | 2023-03-24 | Method for generating bug identification image data, and bug identification system |
| CA3255352A CA3255352A1 (en) | 2022-03-31 | 2023-03-24 | IMAGE DATA GENERATION METHOD FOR INSECT IDENTIFICATION AND INSECT IDENTIFICATION SYSTEM |
Applications Claiming Priority (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| JP2022058795A JP7458089B2 (ja) | 2022-03-31 | 2022-03-31 | 虫同定用画像データの生成方法及び虫同定システム |
| JP2022-058795 | 2022-03-31 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| WO2023190096A1 true WO2023190096A1 (ja) | 2023-10-05 |
Family
ID=88202185
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| PCT/JP2023/011675 Ceased WO2023190096A1 (ja) | 2022-03-31 | 2023-03-24 | 虫同定用画像データの生成方法及び虫同定システム |
Country Status (8)
| Country | Link |
|---|---|
| US (1) | US20260024371A1 (ja) |
| EP (1) | EP4483713A4 (ja) |
| JP (3) | JP7458089B2 (ja) |
| KR (1) | KR20240170915A (ja) |
| AU (1) | AU2023242975A1 (ja) |
| CA (1) | CA3255352A1 (ja) |
| TW (1) | TW202347252A (ja) |
| WO (1) | WO2023190096A1 (ja) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008242712A (ja) * | 2007-03-27 | 2008-10-09 | Olympus Imaging Corp | 画像認識装置および画像認識方法 |
| JP2014142833A (ja) | 2013-01-24 | 2014-08-07 | Earth Kankyo Service Kk | 捕獲虫類の同定方法及び同定システム |
| CN106886758A (zh) * | 2017-01-20 | 2017-06-23 | 北京农业信息技术研究中心 | 基于三维姿态估计的昆虫识别装置及方法 |
| JP2018197940A (ja) * | 2017-05-23 | 2018-12-13 | アース環境サービス株式会社 | 捕獲虫類の同定方法 |
| JP2019016106A (ja) * | 2017-07-05 | 2019-01-31 | 富士通株式会社 | 情報処理プログラム、情報処理装置、情報処理方法、及び情報処理システム |
| JP2020182045A (ja) * | 2019-04-24 | 2020-11-05 | 日本電信電話株式会社 | パノラマ映像合成装置、パノラマ映像合成方法、及びパノラマ映像合成プログラム |
| JP2022058795A (ja) | 2016-12-14 | 2022-04-12 | パナソニックIpマネジメント株式会社 | 食器洗い機 |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2002034034A (ja) * | 2000-07-13 | 2002-01-31 | Yuasa Trading Co Ltd | 画像異常判定機構 |
| JP4200171B2 (ja) * | 2006-10-19 | 2008-12-24 | ニューリー株式会社 | 捕虫シートの検査装置及び捕虫シートにおける捕虫の画像計数方法 |
| JP4256441B2 (ja) * | 2007-09-21 | 2009-04-22 | ニューリー株式会社 | 捕虫装置及び捕虫シートの検査方法 |
| ES2956102T3 (es) * | 2016-10-28 | 2023-12-13 | Verily Life Sciences Llc | Modelos predictivos para clasificar visualmente insectos |
| WO2020172235A1 (en) * | 2019-02-22 | 2020-08-27 | The Johns Hopkins University | Insect specimen analysis system |
-
2022
- 2022-03-31 JP JP2022058795A patent/JP7458089B2/ja active Active
-
2023
- 2023-03-24 US US18/849,274 patent/US20260024371A1/en active Pending
- 2023-03-24 EP EP23780106.3A patent/EP4483713A4/en active Pending
- 2023-03-24 WO PCT/JP2023/011675 patent/WO2023190096A1/ja not_active Ceased
- 2023-03-24 AU AU2023242975A patent/AU2023242975A1/en active Pending
- 2023-03-24 CA CA3255352A patent/CA3255352A1/en active Pending
- 2023-03-24 KR KR1020247032942A patent/KR20240170915A/ko active Pending
- 2023-03-29 TW TW112112076A patent/TW202347252A/zh unknown
-
2024
- 2024-03-11 JP JP2024037408A patent/JP7811402B2/ja active Active
-
2026
- 2026-01-19 JP JP2026006534A patent/JP2026063238A/ja active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008242712A (ja) * | 2007-03-27 | 2008-10-09 | Olympus Imaging Corp | 画像認識装置および画像認識方法 |
| JP2014142833A (ja) | 2013-01-24 | 2014-08-07 | Earth Kankyo Service Kk | 捕獲虫類の同定方法及び同定システム |
| JP2022058795A (ja) | 2016-12-14 | 2022-04-12 | パナソニックIpマネジメント株式会社 | 食器洗い機 |
| CN106886758A (zh) * | 2017-01-20 | 2017-06-23 | 北京农业信息技术研究中心 | 基于三维姿态估计的昆虫识别装置及方法 |
| JP2018197940A (ja) * | 2017-05-23 | 2018-12-13 | アース環境サービス株式会社 | 捕獲虫類の同定方法 |
| JP2019016106A (ja) * | 2017-07-05 | 2019-01-31 | 富士通株式会社 | 情報処理プログラム、情報処理装置、情報処理方法、及び情報処理システム |
| JP2020182045A (ja) * | 2019-04-24 | 2020-11-05 | 日本電信電話株式会社 | パノラマ映像合成装置、パノラマ映像合成方法、及びパノラマ映像合成プログラム |
Non-Patent Citations (1)
| Title |
|---|
| See also references of EP4483713A4 |
Also Published As
| Publication number | Publication date |
|---|---|
| AU2023242975A1 (en) | 2024-09-26 |
| US20260024371A1 (en) | 2026-01-22 |
| TW202347252A (zh) | 2023-12-01 |
| CA3255352A1 (en) | 2025-03-12 |
| JP7811402B2 (ja) | 2026-02-05 |
| EP4483713A4 (en) | 2025-08-20 |
| JP2024060036A (ja) | 2024-05-01 |
| JP2023149958A (ja) | 2023-10-16 |
| EP4483713A1 (en) | 2025-01-01 |
| JP7458089B2 (ja) | 2024-03-29 |
| KR20240170915A (ko) | 2024-12-05 |
| JP2026063238A (ja) | 2026-04-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Staller et al. | Identifying predators at northern bobwhite nests | |
| Schaefer | Video monitoring of shrub-nests reveals nest predators | |
| Vale | Development of baits for tsetse flies (Diptera: Glossinidae) in Zimbabwe | |
| US20190166823A1 (en) | Selective Action Animal Trap | |
| AU2021101929A4 (en) | Pest Management System | |
| US20230206626A1 (en) | Plant disease and pest control method using spectral remote sensing and artificial intelligence | |
| Jack-McCollough et al. | Honeybees tune excitatory and inhibitory recruitment signalling to resource value and predation risk | |
| Ulrich et al. | Ability of bed bug (H emiptera: C imicidae) defensive secretions (E)‐2‐hexenal and (E)‐2‐octenal to attract adults of the common bed bug C imex lectularius | |
| CN109496622A (zh) | 害虫的识别方法及装置、施药信息的确定方法、植保系统 | |
| US12458000B2 (en) | System, method, and apparatus for active pest denial, apiary monitoring and management | |
| KR20230159811A (ko) | 무산쇠족제비 안전 포획장치 | |
| Robertson et al. | Trojan females and Judas goats: evolutionary traps as tools in wildlife management | |
| JP7458089B2 (ja) | 虫同定用画像データの生成方法及び虫同定システム | |
| Linnell et al. | Predator–prey interactions in the canopy | |
| Ellison et al. | Nest defense-grassland bird responses to snakes | |
| US11839878B2 (en) | System and method for sex sorting of mosquitoes | |
| Nielson | Sex pheromone traps: a breakthrough in controlling borers of ornamental trees and shrubs | |
| Khalid et al. | Whispers in the air: Designing acoustic classifiers to detect fruit flies from afar | |
| JP2021193984A (ja) | 家畜への鳥獣害・害虫対策システム | |
| Aman et al. | Modern Technologies of Insect Trapping | |
| KR20210101382A (ko) | 야생동물 퇴치 시스템 및 그 방법 | |
| Jacobs | Predators for Free-Ranging Poultry | |
| Gingrich et al. | Pest birds: biology and management at food processing facilities | |
| KR102531690B1 (ko) | Oled 및 빅데이터와 파장을 기반으로 하는 겹눈해충 퇴치시스템 | |
| HK40106805A (zh) | 有害动物管理系统 |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| 121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 23780106 Country of ref document: EP Kind code of ref document: A1 |
|
| WWE | Wipo information: entry into national phase |
Ref document number: AU2023242975 Country of ref document: AU |
|
| ENP | Entry into the national phase |
Ref document number: 2023242975 Country of ref document: AU Date of ref document: 20230324 Kind code of ref document: A |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 2023780106 Country of ref document: EP |
|
| ENP | Entry into the national phase |
Ref document number: 2023780106 Country of ref document: EP Effective date: 20240926 |
|
| NENP | Non-entry into the national phase |
Ref country code: DE |
|
| WWE | Wipo information: entry into national phase |
Ref document number: 11202406518P Country of ref document: SG |