WO2021136509A1 - 检测包裹的方法、装置、计算设备、物流系统及存储介质 - Google Patents

检测包裹的方法、装置、计算设备、物流系统及存储介质 Download PDF

Info

Publication number
WO2021136509A1
WO2021136509A1 PCT/CN2020/142175 CN2020142175W WO2021136509A1 WO 2021136509 A1 WO2021136509 A1 WO 2021136509A1 CN 2020142175 W CN2020142175 W CN 2020142175W WO 2021136509 A1 WO2021136509 A1 WO 2021136509A1
Authority
WO
WIPO (PCT)
Prior art keywords
package
cloud data
point cloud
volume
regularity
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
Application number
PCT/CN2020/142175
Other languages
English (en)
French (fr)
Inventor
胡茂邦
顾睿
周杨
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Hikrobot Technology Co Ltd
Original Assignee
Hangzhou Hikrobot Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hangzhou Hikrobot Technology Co Ltd filed Critical Hangzhou Hikrobot Technology Co Ltd
Priority to EP20909919.1A priority Critical patent/EP4086829A4/en
Priority to KR1020227024401A priority patent/KR102757317B1/ko
Priority to JP2022539768A priority patent/JP7417745B2/ja
Publication of WO2021136509A1 publication Critical patent/WO2021136509A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • G06Q10/0832—Special goods or special handling procedures, e.g. handling of hazardous or fragile goods
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/30—Measuring arrangements characterised by the use of optical techniques for measuring roughness or irregularity of surfaces
    • G01B11/306—Measuring arrangements characterised by the use of optical techniques for measuring roughness or irregularity of surfaces for measuring evenness
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06395—Quality analysis or management
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/08—Logistics, e.g. warehousing, loading or distribution; Inventory or stock management
    • G06Q10/083—Shipping
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
    • G01B11/00—Measuring arrangements characterised by the use of optical techniques
    • G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
    • G01B11/25—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures by projecting a pattern, e.g. one or more lines, moiré fringes on the object
    • 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
    • G06T2207/10028—Range image; Depth image; 3D point clouds
    • 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
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30108—Industrial image inspection
    • G06T2207/30112—Baggage; Luggage; Suitcase
    • 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
    • Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
    • Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
    • Y02P90/30—Computing systems specially adapted for manufacturing

Definitions

  • This application relates to the field of logistics automation technology, in particular to methods, devices, computing equipment, logistics systems, and storage media for detecting packages.
  • the regularity indicates the degree of surface flatness of the package and the closeness of the loaded volume of the package to the upper limit of the capacity.
  • This application proposes methods, devices, computing equipment, logistics systems, and storage media for package detection, which can realize automatic detection of package regularity, thereby improving package detection efficiency and consistency of detection.
  • a method for detecting packages including:
  • the regularity of the package is determined according to the appearance characteristics of the package, wherein the regularity represents the surface smoothness of the package and/or the closeness of the loaded amount of the package to the upper limit of the package capacity.
  • the acquiring point cloud data on the outer surface of the package includes:
  • the point cloud data is determined.
  • the determining the shape feature of the package according to the point cloud data includes at least one of the following steps:
  • volume duty cycle is the ratio of the volume of the package to the volume of the smallest circumscribed cuboid of the package
  • the contour rectangle degree being the ratio of the area of the horizontal projection area of the package to the area of the smallest circumscribed rectangular area of the horizontal projection area.
  • the determining the regularity of the package according to the shape feature of the package includes:
  • the determining the flatness of the top surface of the package according to the point cloud data includes:
  • the ratio of the number of points in the target subset to the number of points in the point cloud data is used as the top surface flatness.
  • the determining the main direction of the point cloud data includes:
  • the determining the volume duty cycle of the package according to the point cloud data includes:
  • the volume duty ratio is calculated according to the volume of the cuboid and the volume of the package.
  • the determining the volume of the package according to the point cloud data includes:
  • the sum of the volumes corresponding to multiple grids is used as the volume of the package.
  • the determining the outline rectangularity of the package according to the point cloud data includes:
  • the ratio of the area of the horizontal projection area to the area of the rectangular area is taken as the outline rectangularity.
  • the determining the main direction of the point cloud data includes:
  • a device for detecting packages including:
  • the acquiring unit acquires the point cloud data of the outer surface of the package
  • a feature extraction unit which determines the shape feature of the package according to the point cloud data
  • the regularity determining unit determines the regularity of the package according to the appearance characteristics of the package, wherein the regularity represents the surface smoothness of the package and the closeness of the loaded amount of the package to the upper limit of the capacity.
  • a computing device including: a memory; a processor; a program, stored in the memory and configured to be executed by the processor, the program including a package for executing the inspection package according to the present application Method of instruction.
  • a storage medium storing a program, the program including instructions that, when executed by a computing device, cause the computing device to execute the method for detecting packages according to the present application.
  • a logistics system including: computing equipment; a conveyor belt; and a point cloud data collection device for collecting point cloud data of packages on the conveyor belt.
  • a computer program product containing instructions, which when the computer program product runs on a computer, causes the computer to execute the package detection method of the present application.
  • the shape characteristics of the package can be obtained, and the regularity can be determined according to the shape characteristics.
  • the package detection scheme can avoid the trouble of manually detecting the package regularity, and realize the automatic detection of the package regularity by the equipment, thereby improving the efficiency of package regularity detection and the consistency of the detection results.
  • Figure 1 shows a schematic diagram of a logistics system according to some embodiments of the present application
  • FIG. 2A shows a flowchart of a method 200 for detecting a package according to some embodiments of the present application
  • Figure 2B shows a schematic diagram of a package according to some embodiments of the present application.
  • FIG. 2C shows a schematic diagram of point cloud data according to some embodiments of the present application.
  • FIG. 3 shows a flowchart of a method 300 for determining the appearance characteristics of a package according to some embodiments of the present application
  • FIG. 4 shows a flowchart of a method 400 for detecting a package according to some embodiments of the present application
  • FIG. 5A shows a flowchart of a method 500 for determining the flatness of the top surface of a package according to some embodiments of the present application
  • FIG. 5B shows a flowchart of determining the main direction of point cloud data according to some embodiments of the present application
  • FIG. 6 shows a flowchart of a method 600 for determining the flatness of the top surface of a package according to some embodiments of the present application
  • FIG. 7 shows a flowchart of a method 700 of a volume duty cycle according to some embodiments of the present application
  • FIG. 8 shows a flowchart of a method 800 for determining a package volume according to some embodiments of the present application
  • FIG. 9 shows a flow chart of a method 900 for outline rectangularity according to some embodiments of the present application.
  • FIG. 10 shows a schematic diagram of an apparatus 1000 for detecting packages according to some embodiments of the present application.
  • FIG. 11 shows a schematic diagram of an apparatus 1100 for detecting packages according to some embodiments of the present application.
  • Figure 12 shows a schematic diagram of a computing device according to some embodiments of the present application.
  • Fig. 1 shows a schematic diagram of a logistics system according to some embodiments of the present application.
  • the logistics system includes a conveyor belt 110, a point cloud data collection device 120 and a computing device 130.
  • the conveyor belt 110 conveys the packages according to the conveying direction.
  • the package 140 shown in FIG. 1 is delivered.
  • the conveying direction is the direction from left to right in FIG. 1.
  • the point cloud data acquisition device 120 may be, for example, a laser radar or a depth camera.
  • the depth camera is, for example, a structured light camera, a time of flight (Time of Flight, abbreviated as ToF) camera, a binocular vision (Stereo) camera, and the like.
  • ToF Time of Flight
  • Stepo binocular vision
  • the point cloud data collection device 120 can scan the package passing through the conveyor belt and located in the detection area S1, and output the scanning result to the computing device 130.
  • the computing device 130 may be, for example, a server, a notebook computer, a tablet computer, or a handheld business communication device.
  • the computing device 130 may determine the point cloud data on the outer surface of the package according to the scanning result from the point cloud data collection device 120.
  • Point cloud data is a collection of points on the outer surface of the package.
  • the coordinates of each point in the point cloud data can be expressed as coordinates in the coordinate system of the point cloud data acquisition device 120, or as coordinates in the world coordinate system.
  • the set of points on the surface of which the above-mentioned point cloud data is specifically packaged depends on the installation position and angle of the point cloud data collection device 120.
  • the point cloud data is the collection of points on the outer surface of the package that can be collected by the point cloud data collection device 120.
  • the point cloud data collection device 120 is located above the conveyor belt 110 and scans the package in a direction perpendicular to the conveyor belt 110, then the point cloud data is a collection of points on the upper surface of the package and points on the side protrusions, that is, point cloud The collection of points on the outer surface of the package that can be collected by the data collection device 120.
  • the computing device 130 can automatically detect the regularity of the package based on the point cloud data of the package.
  • the regularity indicates the degree of surface flatness of the package and the closeness of the loaded amount of the package to the upper limit of the capacity.
  • FIG. 2A shows a flowchart of a method 200 for detecting a package according to some embodiments of the present application.
  • the method 200 is executed by the computing device 130, for example.
  • step S201 point cloud data on the outer surface of the package is acquired.
  • the computing device may obtain point cloud data on the upper surface of the package.
  • the upper surface of the package refers to the surface area of the package that can be viewed from a top view.
  • the computing device may receive the scanning result obtained by scanning the package on the conveyor belt by the data collection device. According to the scanning result, the computing device can determine the point cloud data of the outer surface of the package in step S201. Each point in the point cloud data can be considered to correspond to a point on the outer surface of the package. The information of each point in the point cloud data includes the position information of the point. The position information of a point can be expressed as coordinates in a three-dimensional coordinate system, for example.
  • FIG. 2B shows a schematic diagram of a package according to some embodiments of the present application.
  • Fig. 2C shows a schematic diagram of point cloud data according to some embodiments of the present application. The point cloud data in Fig. 2C can reflect the appearance characteristics of the package in Fig. 2B.
  • the shape feature of the package is determined according to the point cloud data.
  • the shape feature may include, for example, at least one of top surface flatness, volume duty ratio, ideal volume duty ratio, and outline rectangularity.
  • the regularity of the package is determined according to the appearance characteristics of the package.
  • the regularity indicates the degree of surface flatness of the package and/or the closeness of the loaded amount of the package to the upper limit of the package capacity.
  • the regularity is the weighted sum of various shape features of the package.
  • the computing device may calculate the weighted sum of at least one of the top surface flatness, volume duty ratio, ideal volume ratio, and outline rectangularity, and use the weighted sum as the regularity of the package.
  • the shape feature of the package can be obtained, and the regularity can be determined according to the shape feature.
  • the method 200 can avoid the trouble of manually detecting the regularity of the package, and realize the automatic detection of the regularity of the package by the device, thereby improving the efficiency of package regularity detection and the consistency of the detection result.
  • step S202 may be implemented as method 300. That is to say, the above step S202 may include the following steps:
  • step S301 the flatness of the top surface of the package is determined according to the point cloud data.
  • the flatness of the top surface is used to characterize the smoothness of the top surface of the package.
  • the top surface may be the surface area of the package that can be viewed from a top view.
  • step S302 the volume duty ratio of the package is determined according to the point cloud data.
  • the volume duty cycle is the ratio of the volume of the package to the volume of the smallest circumscribed cuboid.
  • the computing device can determine the volume of the package according to the coordinates of each point corresponding to the package represented by the point cloud data. It is also possible to determine the minimum circumscribed cuboid of the package according to the coordinates of each point corresponding to the package indicated by the point cloud data, and then determine the volume duty cycle according to the ratio of the volume of the package to the volume of the minimum circumscribed cuboid of the package.
  • step S303 according to the point cloud data, the ideal volume percentage of the package is determined.
  • the ideal volume ratio is the ratio of the volume of the package to the upper limit of the volume of the package.
  • the higher the ideal volume ratio the closer the parcel loading capacity is to the upper limit of the parcel's capacity.
  • the volume of the package can be determined according to the coordinates of each point corresponding to the package represented by the point cloud data.
  • the upper limit of the capacity of the package can be obtained and stored in advance.
  • the computing device may obtain the capacity of the package in advance and store it, so as to calculate the ideal volume ratio of the package.
  • the computing device may obtain the capacity of each package in advance, and store it corresponding to the identifier of each package.
  • the above-mentioned point cloud data collection device 120 scans the package, it can also obtain the identifier of the package by scanning the two-dimensional code on the package, and then send it to the computing device. In this way, the computing device can determine the capacity of the corresponding package according to the identifier, and then determine the ideal volume ratio of the package.
  • step S304 the contour rectangle degree of the package is determined according to the point cloud data.
  • the contour rectangle degree is the ratio of the area of the enclosed horizontal projection area to the area of the smallest circumscribed rectangular area of the horizontal projection area. That is, the ratio of the area of the projection area wrapped in the horizontal plane to the area of the smallest circumscribed rectangular area of the projection area on the horizontal plane.
  • the higher the rectangular degree of the outline the closer the shape of the package is to the cuboid or cube, and the more regular the shape of the package.
  • the computing device can determine the projection area of the package on the horizontal plane based on the point cloud data of the package, and then can determine the minimum circumscribed rectangle of the projection area, and then can calculate the area of the packaged horizontal projection area and the horizontal projection The ratio of the area of the smallest circumscribed rectangular area of the area determines the rectangularity of the package outline.
  • the method 300 can analyze the shape characteristics of the package from a variety of angles, so that a variety of shape characteristics can be used to determine the regularity of the package. Since the shape features can include top surface flatness, volume duty ratio, ideal volume ratio and/or outline rectangularity, the calculated package regularity can indicate the surface flatness of the package and/or the amount of the package loaded and the package The proximity of the upper capacity limit.
  • FIG. 4 shows a flowchart of a method 400 for detecting a package according to some embodiments of the present application.
  • the method 400 is executed by the computing device 130, for example.
  • step S401 point cloud data on the outer surface of the package is acquired.
  • step S402 the shape feature of the package is determined according to the point cloud data.
  • step S403 the regularity of the package is determined according to the appearance characteristics of the package.
  • step S401 to step S403 are the same as the above step S201 to step S203 respectively, please refer to the description of the above step S201 to step S203, which will not be repeated here.
  • step S404 it is determined whether the regularity reaches the regularity threshold.
  • the regularity threshold can be set in advance according to actual needs. When the regularity requirement of the package is high, the regularity threshold can be higher; when the regularity requirement of the package is low, the regularity threshold can be lower.
  • step S404 When it is determined in step S404 that the regularity reaches the regularity threshold, in the method 400, continue to perform step S405 to determine that the classification result of the package is the first category. When it is determined in step S404 that the regularity is lower than the regularity threshold, step S406 is continued in the method 400 to determine that the classification result of the package is the second category.
  • the first category is the type that meets the regularity requirements
  • the second category is the type that does not meet the regularity requirements.
  • step S407 can be continued.
  • the classification result of the package is sent to the sorting device.
  • the sorting device can sort the packages according to the sorting results.
  • the sorting device can sort the packages belonging to the second category on the conveyor belt to the target location.
  • the staff can sort the packages of the second category, so that the packages can meet the regularity requirements after sorting, thereby improving the transportation convenience of the packages.
  • the category of the package is the first category, it means that the regularity of the package is high and meets the regularity requirements for the package, so the next step can be processed, for example, the package can be loaded into the car, warehoused, or Enter the next sorting process, etc., without the need for staff to sort it out.
  • step S301 may be implemented as method 500.
  • the above step S301 may include the following steps:
  • step S501 the main direction of the point cloud data is determined.
  • the main direction is used to reflect the overall orientation of the point cloud data.
  • step S501 may be implemented as a flowchart shown in FIG. 5B. That is, step S501 may include the following steps:
  • step S5011 the center point of the point cloud data is calculated.
  • the coordinates of the center point are the mean value of the coordinates of the points in the point cloud data.
  • step S5012 the coordinates of each point in the point cloud data are subtracted from the coordinates of the center point to obtain the point cloud data with the center offset.
  • step S5013 the first matrix and the second matrix composed of the point cloud data with the center shifted are determined.
  • the second matrix is the transposed matrix of the first matrix.
  • step S5014 the product of the first matrix and the second matrix is calculated to obtain the covariance matrix.
  • the first matrix is assumed to be A
  • the second matrix is A T.
  • the covariance matrix cov AA T.
  • step S5015 Singular Value Decomposition (SVD) is performed on the covariance matrix to obtain the eigenvector of the covariance matrix, and then the eigenvector with the largest eigenvalue is taken as the main direction of the point cloud data.
  • SVD Singular Value Decomposition
  • step S502 a target subset of the point cloud data is determined.
  • the angle between the normal direction and the main direction of each point in the target subset is less than the angle threshold.
  • the included angle threshold is, for example, 10 degrees, 15 degrees, 18 degrees, etc., but it is not limited thereto.
  • the computing device can calculate the angle between the normal direction of each point in the point cloud data and the main direction, and then determine the set of points whose angle is less than the angle threshold as a point Target subset of cloud data.
  • the normal of each point is a straight line that passes through the point and is perpendicular to the local plane where the point is located.
  • step S503 the ratio of the number of points in the target subset to the number of points in the point cloud data is used as the top surface flatness.
  • the higher the flatness of the top surface the smoother the top surface of the package.
  • the greater the number of points in the target subset the greater the number of points whose angle between the corresponding normal direction and the main direction is less than the included angle threshold, and the greater the number of points on the top of the package on a plane. More, so it can be determined that the top surface of the package is smoother.
  • step S301 may be implemented as method 600.
  • the above step S301 may include the following steps:
  • step S601 the point cloud data is down-sampled to obtain sampled point cloud data.
  • the data scale of the point cloud data can be reduced, thereby improving the efficiency of calculating the flatness of the top surface.
  • step S602 based on the principal component analysis algorithm, the principal direction of the sampled point cloud data is determined.
  • the main direction is used to reflect the overall orientation of the point cloud data.
  • the principal component analysis algorithm may specifically be the above-mentioned singular value decomposition algorithm, which will not be repeated here.
  • step S603 a target subset of the sampled point cloud data is determined.
  • the angle between the normal direction and the main direction of each point in the target subset is less than the angle threshold.
  • the included angle threshold is, for example, 10 degrees, but it is not limited to this. Since step S603 is the same as the above step S502, please refer to the description of the above step S502, which will not be repeated here.
  • step S604 the ratio of the number of points in the target subset to the number of points in the sampled point cloud data is used as the top surface flatness.
  • the flatness of the top surface can be obtained according to the following formula.
  • flat_ratio is the flatness of the top surface
  • N angle is the number of points in the target subset
  • N sampled represents the number of points in the sampled point cloud data.
  • step S302 may be implemented as method 700. That is, step S302 may include the following steps:
  • step S701 the shape model of the package is determined according to the point cloud data.
  • step S701 can determine the three-dimensional model of the package according to the point cloud data. Since the point cloud data can represent the positions of various points outside the package, a three-dimensional model of the package can be constructed based on the point cloud data.
  • step S702 the minimum circumscribed cuboid of the package is determined according to the shape model of the package, and the volume of the cuboid is determined.
  • the minimum circumscribed cuboid of the package can be determined based on the shape model, and then the volume of the minimum circumscribed cuboid can be calculated.
  • step S703 the volume of the package is determined according to the point cloud data.
  • various volume calculation methods may be used in step S703 to determine the volume of the package.
  • step S703 may be implemented as method 800. That is, step S703 may include the following steps:
  • step S801 point cloud data is projected into a predetermined plane to obtain projection points corresponding to the point cloud data.
  • the predetermined plane may be the carrying surface of the package. For example, when the package is placed on a conveyor belt, since the lower surface of the package coincides with the surface of the conveyor belt, the predetermined plane may be the surface of the conveyor belt.
  • step S802 rasterization processing is performed on the projection points corresponding to the point cloud data to obtain multiple grids containing the projection points.
  • step S803 the volume corresponding to each grid is calculated, and the volume corresponding to each grid is the product of the area of the grid and the height of the point cloud data projected on the grid.
  • the height of the point cloud data projected to the grid is the mean value of the height of the point cloud data projected to the grid or the height value corresponding to the largest number of points within the height range of the point cloud data projected to the grid.
  • the height value corresponding to the largest number of points can also be understood as: after the point cloud data projected to the grid is grouped according to height, the height value corresponding to the group that includes the largest number of points.
  • the height of a single point in the point cloud data is the difference between the vertical coordinate value of the point in the three-dimensional coordinate system and the vertical coordinate value of the conveyor belt.
  • the volume corresponding to each grid can be understood as the volume of the cylinder corresponding to the grid.
  • the column can be considered as the column obtained by vertically splitting the wrapped shape model by the grid.
  • step S804 the sum of the volumes corresponding to the multiple grids is used as the volume of the package.
  • step S804 the sum of the volumes corresponding to the multiple grids is regarded as the volume of the package, that is, the sum of the volumes corresponding to all the grids is regarded as the volume of the package.
  • the method 800 can calculate the integral volume of the package by rasterizing the point cloud data. Therefore, even if the surface of the package is not flat enough, the method 800 can determine the volume of the package more accurately.
  • step S704 the volume duty ratio is calculated according to the volume of the cuboid and the volume of the package.
  • the volume duty ratio can be calculated according to the following formula.
  • vol integral is the volume of the package
  • vol box is the volume of the smallest circumscribed cuboid of the package.
  • step S304 may be implemented as method 900. That is to say, the above step S304 may include the following steps:
  • step S901 the horizontal projection area of the point cloud data in the horizontal plane is determined, and the area of the horizontal projection area is determined.
  • step S902 the minimum circumscribed rectangular area of the horizontal projection area is determined, and the area of the rectangular area is determined.
  • step S903 the ratio of the area of the horizontal projection area to the area of the rectangular area is taken as the outline rectangle degree.
  • the outline rectangle degree may be calculated according to the following manner.
  • rectangularity is the degree of outline rectangle
  • S convex_hull is the area of the horizontal projection area
  • S min_rect is the area of the smallest circumscribed rectangular area. The closer the contour rectangle degree is to 1, the closer the top surface of the package is to a rectangle, and the better the package regularity.
  • the regularity of the package is related to the flatness of the top surface, the volume duty ratio, the ideal volume ratio, and the contour rectangularity.
  • FIG. 10 shows an apparatus 1000 for detecting packages according to some embodiments of the present application.
  • the apparatus 1000 may be deployed in the computing device 130, for example.
  • the device 1000 includes an acquisition unit 1001, a feature extraction unit 1002, and a regularity determination unit 1003.
  • the obtaining unit 1001 is used to obtain point cloud data on the outer surface of the package.
  • the feature extraction unit 1002 is used to determine the shape feature of the package according to the point cloud data.
  • the regularity determining unit 1003 is used to determine the regularity of the package according to the shape characteristics of the package. Among them, the regularity indicates the degree of surface flatness of the package and/or the closeness of the loaded amount of the package to the upper limit of the package capacity.
  • a more specific implementation manner of the apparatus 1000 is the same as that of the method 200, and will not be repeated here.
  • the shape feature of the package can be obtained, and the regularity can be determined according to the shape feature.
  • the device 1000 can avoid the trouble of manually detecting the regularity of the package, and realize the automatic detection of the regularity of the package by the device, thereby improving the efficiency of package regularity detection and the consistency of the detection result.
  • FIG. 11 shows an apparatus 1100 for detecting a package according to some embodiments of the present application.
  • the apparatus 1100 may be deployed in the computing device 130, for example.
  • the device 1100 includes an acquisition unit 1101, a feature extraction unit 1102, a regularity determination unit 1103, and a classification unit 1104.
  • the obtaining unit 1101 may be used to obtain point cloud data on the outer surface of the package.
  • the feature extraction unit 1102 can be used to determine the shape feature of the package according to the point cloud data.
  • the regularity determining unit 1103 may be used to determine the regularity of the package according to the appearance characteristics of the package.
  • the classification unit 1104 may be used to determine whether the regularity reaches the regularity threshold. When the regularity reaches the regularity threshold, the classification unit 1104 determines that the classification result of the package is the first category. When the regularity is lower than the regularity threshold, the classification unit 1104 determines that the classification result of the package is the second category.
  • the obtaining unit 1101 may receive the scanning result obtained by scanning the package on the conveyor belt by the data collecting device. According to the scanning result, the acquiring unit 1101 determines the point cloud data.
  • the feature extraction unit 1102 can determine one or more shape features of the package. For example, based on the point cloud data, the feature extraction unit 1102 determines the flatness of the top surface of the package. Among them, the flatness of the top surface is used to characterize the smoothness of the top surface of the package. According to the point cloud data, the feature extraction unit 1102 determines the volume duty ratio of the package. The volume duty cycle is the ratio of the volume of the package to the volume of the smallest circumscribed cuboid. According to the point cloud data, the feature extraction unit 1102 determines the ideal volume ratio of the package. The ideal volume ratio is the ratio of the volume of the package to the upper limit of the volume of the package. According to the point cloud data, the rectangularity of the package outline is determined. The contour rectangle degree is the ratio of the area of the enclosed horizontal projection area to the area of the smallest circumscribed rectangular area of the horizontal projection area.
  • the regularity determining unit 1103 can be used to calculate the weighted sum of at least one of the top surface flatness, volume duty ratio, ideal volume ratio, and contour rectangularity, and use the weighted sum as the package Regularity.
  • the feature extraction unit 1102 may be used to determine the main direction of the point cloud data.
  • the main direction is used to reflect the overall orientation of the point cloud data.
  • the feature extraction unit 1102 can also be used to determine a target subset of the point cloud data. The angle between the normal direction and the main direction of each point in the target subset is less than the angle threshold. In this way, the feature extraction unit 1102 can use the ratio of the number of points in the target subset to the number of points in the point cloud data as the top surface flatness.
  • the feature extraction unit 1102 may also be used to perform a down-sampling operation on the point cloud data to obtain sampled point cloud data.
  • the feature extraction unit 1102 may be used to determine the shape model of the package according to the point cloud data. According to the shape model of the package, the feature extraction unit 1102 can be used to determine the smallest circumscribed cuboid of the package and determine the volume of the cuboid. According to the point cloud data, the feature extraction unit 1102 can determine the volume of the package. According to the volume of the cuboid and the volume of the package, the feature extraction unit 1102 can calculate the volume duty ratio.
  • the feature extraction unit 1102 can be used to rasterize the point cloud data according to a rasterized array in a predetermined plane to obtain points corresponding to multiple grids, where each grid corresponds to The horizontal projection of the point belongs to the range of the grid. According to the points corresponding to each grid, the feature extraction unit 1102 calculates the volume corresponding to the grid. Wherein, the volume corresponding to the grid is the product of the area of the grid and the height of the point corresponding to the grid. The feature extraction unit 1102 uses the sum of the volumes corresponding to the multiple grids as the volume of the package.
  • the feature extraction unit 1102 can be used to determine the horizontal projection area of the point cloud data in the horizontal plane, and determine the area of the horizontal projection area.
  • the feature extraction unit 1102 may determine the minimum circumscribed rectangular area of the horizontal projection area, and determine the area of the rectangular area.
  • the feature extraction unit 1102 may use the ratio of the area of the horizontal projection area to the area of the rectangular area as the contour rectangle degree.
  • the feature extraction unit 1102 may be used to calculate the center point of the point cloud data.
  • the coordinates of the center point are the mean value of the coordinates of the points in the point cloud data.
  • the feature extraction unit 1102 subtracts the coordinates of the center point from the coordinates of each point in the point cloud data to obtain the point cloud data with the center offset.
  • the feature extraction unit 1102 determines the first matrix and the second matrix composed of the point cloud data with the center shifted.
  • the second matrix is the transposed matrix of the first matrix.
  • the feature extraction unit 1102 obtains the covariance matrix by calculating the product of the first matrix and the second matrix.
  • the feature extraction unit 1102 performs singular value decomposition on the covariance matrix, and takes the feature vector with the largest feature value as the main direction of the point cloud data.
  • FIG. 12 shows a schematic diagram of a computing device according to some embodiments of the present application.
  • the computing device includes one or more processors (Central Processing Unit, CPU) 1202, a communication module 1204, a memory 1206, a user interface 1210, and a communication bus 1208 for interconnecting these components.
  • processors Central Processing Unit, CPU
  • communication module 1204 a memory 1206, a user interface 1210, and a communication bus 1208 for interconnecting these components.
  • FIG. 12 only one processor is shown in FIG. 12 as an illustration.
  • the processor 1202 may receive and send data through the communication module 1204 to implement network communication and/or local communication.
  • the user interface 1210 includes one or more output devices 1212, which includes one or more speakers and/or one or more visual displays.
  • the user interface 1210 also includes one or more input devices 1214.
  • the user interface 1210 can receive instructions from a remote controller, for example, but is not limited thereto. Among them, only one output device and one input device are shown in FIG. 12 as an illustration.
  • the memory 1206 may be a high-speed random access memory, such as DRAM, SRAM, DDR RAM, or other random access solid-state storage devices; or a non-volatile memory, such as one or more magnetic disk storage devices, optical disk storage devices, or flash memory devices, Or other non-volatile solid-state storage devices.
  • a high-speed random access memory such as DRAM, SRAM, DDR RAM, or other random access solid-state storage devices
  • non-volatile memory such as one or more magnetic disk storage devices, optical disk storage devices, or flash memory devices, Or other non-volatile solid-state storage devices.
  • the memory 1206 stores an instruction set executable by the processor 1202, including:
  • Operating system 1216 including programs for processing various basic system services and performing hardware-related tasks
  • the application 1218 includes various programs for realizing the above detection of packages, for example, it may include devices 1000 and 1100 for detecting packages. This kind of program can realize the processing flow in the above-mentioned examples, for example, it can include the method of detecting the package.
  • the application 1218 may include various programs for realizing the aforementioned package detection, for example, may include a package detection device 1000 or a package detection device 1100.
  • each embodiment of the present application can be implemented by a data processing program executed by a data processing device such as a computer.
  • the data processing program constitutes the present invention.
  • a data processing program usually stored in a storage medium is executed by directly reading the program out of the storage medium or by installing or copying the program to a storage device (such as a hard disk and/or a memory) of the data processing device. Therefore, such a storage medium also constitutes the present invention.
  • the storage medium can use any type of recording method, such as paper storage medium (such as paper tape, etc.), magnetic storage medium (such as floppy disk, hard disk, flash memory, etc.), optical storage medium (such as CD-ROM, etc.), magneto-optical storage medium (such as MO, etc.) and so on.
  • paper storage medium such as paper tape, etc.
  • magnetic storage medium such as floppy disk, hard disk, flash memory, etc.
  • optical storage medium such as CD-ROM, etc.
  • magneto-optical storage medium Such as MO, etc.
  • this application also discloses a non-volatile storage medium in which a program is stored.
  • the program includes instructions that, when executed by the processor, cause the computing device to execute the method of detecting packages according to the present application.
  • the application also discloses a computer program product containing instructions.
  • the computer program product runs in a computer, the computer executes the package detection method described in any of the foregoing embodiments.
  • the computer when the computer program product is running in the computer, the computer can obtain the shape characteristics of the package based on the point cloud data of the package, and determine the regularity according to the shape characteristics.
  • the package detection scheme can avoid the trouble of manually detecting the package regularity, and realize the automatic detection of the package regularity by the equipment, thereby improving the efficiency of package regularity detection and the consistency of the detection results.

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Economics (AREA)
  • Human Resources & Organizations (AREA)
  • Quality & Reliability (AREA)
  • Development Economics (AREA)
  • Strategic Management (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Tourism & Hospitality (AREA)
  • Operations Research (AREA)
  • Geometry (AREA)
  • Educational Administration (AREA)
  • Game Theory and Decision Science (AREA)
  • Length Measuring Devices By Optical Means (AREA)
  • Image Analysis (AREA)

Abstract

提出了检测包裹的方法、装置、计算设备、物流系统及存储介质。其中，一种检测包裹的方法，包括：获取包裹的外表面的点云数据（S201）；根据所述点云数据，确定所述包裹的外形特征（S202）；根据所述包裹的外形特征，确定所述包裹的规整度（S203），其中，所述规整度表示包裹的表面平整程度和/或包裹已装载量与包裹容量上限的接近程度。

Description

检测包裹的方法、装置、计算设备、物流系统及存储介质
本申请要求于2019年12月31日提交中国专利局、申请号为201911413069.6发明名称为“检测包裹的方法、装置、计算设备、物流系统及存储介质”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及物流自动化技术领域，特别涉及检测包裹的方法、装置、计算设备、物流系统及存储介质。
背景技术
目前，在物流的应用场景中，工作人员通常需要对传送带上包裹的规整度进行检测，以便保证运输的包裹有较高的规整度，以便保证运输效率。其中，规整度表示包裹的表面平整程度和包裹已装载量与容量上限的接近程度。
然而，人工检测包裹的规整度的方式效率不高，且检测结果的一致性较差。为此，缺少一种自动化检测包裹的规整度的方案。
发明内容
本申请提出了检测包裹的方法、装置、计算设备、物流系统及存储介质，能够实现包裹规整度的自动化检测，进而提高包裹的检测效率和提高检测的一致性。
根据本申请一个方面，提供一种检测包裹的方法，包括：
获取包裹的外表面的点云数据；
根据所述点云数据，确定所述包裹的外形特征；
根据所述包裹的外形特征，确定所述包裹的规整度，其中，所述规整度表示包裹的表面平整程度和/或包裹已装载量与包裹容量上限的接近程度。
在一些实施例中，所述获取包裹的外表面的点云数据，包括：
接收点云数据采集装置对传送带上所述包裹进行扫描而得到的扫描结果；
根据所述扫描结果，确定所述点云数据。
在一些实施例中，所述根据所述点云数据，确定所述包裹的外形特征，包括下述中至少一个步骤：
根据所述点云数据，确定所述包裹的顶面平整度，其中，所述顶面平整 度用于表征所述包裹顶部表面的平滑程度；
根据所述点云数据，确定所述包裹的体积占空比，所述体积占空比为所述包裹的体积与所述包裹的最小外接长方体的体积之比；
根据所述点云数据，确定所述包裹的理想体积占比，所述理想体积占比为所述包裹的体积与所述包裹的容量上限之比；
根据所述点云数据，确定所述包裹的轮廓矩形度，所述轮廓矩形度为所述包裹的水平投影区域的面积与水平投影区域的最小外接矩形区域的面积之比。
在一些实施例中，所述根据所述包裹的外形特征，确定所述包裹的规整度，包括：
计算所述顶面平整度、体积占空比、理想体积占比和轮廓矩形度中至少一个外形特征的加权和，并将所述加权和作为所述包裹的规整度。
在一些实施例中，所述根据所述点云数据，确定所述包裹的顶面平整度，包括：
确定所述点云数据的主方向，所述主方向用于反映所述点云数据的整体朝向；
确定所述点云数据的目标子集，目标子集中每个点的法向与所述主方向夹角小于夹角阈值；
将所述目标子集的点数量与所述点云数据的点数量之比作为所述顶面平整度。
在一些实施例中，所述确定所述点云数据的主方向，包括：
对点云数据进行下采样操作，得到采样后的点云数据；
确定采样后的点云数据的主方向。
在一些实施例中，所述根据所述点云数据，确定所述包裹的体积占空比，包括：
根据所述点云数据确定包裹的外形模型；
根据所述包裹的外形模型确定所述包裹的最小外接长方体，并确定所述长方体的体积；
根据所述点云数据，确定所述包裹的体积；
根据所述长方体的体积和所述包裹的体积，计算所述体积占空比。
在一些实施例中，所述根据所述点云数据，确定所述包裹的体积，包括：
将所述点云数据投影至预定平面内，得到所述点云数据对应的投影点；
对所述点云数据对应的投影点进行栅格化处理，得到包含投影点的多个栅格；
计算每个栅格对应的体积，每个栅格对应的体积为该栅格的面积与投影至该栅格的点云数据的高度之积；
将多个栅格对应的体积之和作为包裹的体积。
在一些实施例中，所述根据所述点云数据，确定所述包裹的轮廓矩形度，包括：
确定所述点云数据在水平面内的水平投影区域，并确定水平投影区域的面积；
确定所述水平投影区域的最小外接矩形区域，并确定所述矩形区域的面积；
将所述水平投影区域的面积与所述矩形区域的面积之比作为所述轮廓矩形度。
在一些实施例中，所述确定所述点云数据的主方向，包括：
计算点云数据的中心点，所述中心点的坐标为点云数据中点的坐标均值；
将点云数据中各点的坐标减去中心点的坐标，得到中心偏移后的点云数据；
确定中心偏移后的点云数据组成的第一矩阵和第二矩阵，其中第二矩阵为第一矩阵的转置矩阵；
计算第一矩阵和第二矩阵之积，得到协方差矩阵；
对协方差矩阵进行奇异值分解，取特征值最大的特征向量作为点云数据的主方向。
根据本申请一方面，提供一种检测包裹的装置，包括：
获取单元，获取包裹的外表面的点云数据；
特征提取单元，根据所述点云数据，确定所述包裹的外形特征；
规整度确定单元，根据所述包裹的外形特征，确定所述包裹的规整度， 其中，所述规整度表示包裹的表面平整程度和包裹已装载量与容量上限的接近程度。
根据本申请一方面，提供一种计算设备，包括：存储器；处理器；程序，存储在该存储器中并被配置为由所述处理器执行，所述程序包括用于执行根据本申请的检测包裹的方法的指令。
根据本申请一方面，提供一种存储介质，存储有程序，所述程序包括指令，所述指令当由计算设备执行时，使得所述计算设备执行根据本申请的检测包裹的方法。
根据本申请一方面，提供一种物流系统，包括：计算设备；传送带；点云数据采集装置，用于采集所述传送带上包裹的点云数据。
根据本申请一方面，提供一种包含指令的计算机程序产品，当所述计算机程序产品在计算机中运行时，使得所述计算机执行本申请的检测包裹的方法。
综上，根据本申请的检测包裹的方案，基于包裹的点云数据，可以获取包裹的外形特征，并且根据外形特征确定规整度。检测包裹的方案可以避免人工检测包裹规整度的麻烦，实现由设备对包裹进行规整度的自动检测，进而提高包裹规整度检测的效率和检测结果的一致性。
附图说明
为了更清楚地说明本发明实施例和现有技术的技术方案，下面对实施例和现有技术中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1示出了根据本申请一些实施例的物流系统的示意图；
图2A示出了根据本申请一些实施例的检测包裹的方法200的流程图；
图2B示出了根据本申请一些实施例的包裹的示意图；
图2C示出了根据本申请一些实施例的点云数据的示意图；
图3示出了根据本申请一些实施例的确定包裹的外形特征的方法300的流程图；
图4示出了根据本申请一些实施例的检测包裹的方法400的流程图；
图5A示出了根据本申请一些实施例的确定包裹的顶面平整度的方法500的流程图；
图5B示出了根据本申请一些实施例的确定点云数据的主方向的流程图；
图6示出了根据本申请一些实施例的确定包裹的顶面平整度的方法600的流程图；
图7示出了根据本申请一些实施例的体积占空比的方法700的流程图；
图8示出了根据本申请一些实施例的确定包裹体积的方法800的流程图；
图9示出了根据本申请一些实施例的轮廓矩形度的方法900的流程图；
图10示出了根据本申请一些实施例的检测包裹的装置1000的示意图；
图11示出了根据本申请一些实施例的检测包裹的装置1100的示意图；
图12示出了根据本申请一些实施例的计算设备的示意图。
具体实施方式
为使本发明的目的、技术方案、及优点更加清楚明白，以下参照附图并举实施例，对本发明进一步详细说明。显然，所描述的实施例仅仅是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
图1示出了根据本申请一些实施例的物流系统的示意图。如图1所示，物流系统包括传送带110、点云数据采集装置120和计算设备130。
传送带110按照传送方向传送包裹。例如，传送图1示出的包裹140。其中，传送方向如在图1中从左向右的方向。
点云数据采集装置120例如可以是激光雷达或者深度相机。深度相机例如为结构光相机、选用时间飞行(Time of Flight，缩写为ToF)相机、双目视觉(Stereo)相机等。
点云数据采集装置120可以对经过传送带上位于检测区域S1的包裹进行扫描，并向计算设备130输出扫描结果。
计算设备130例如可以是服务器、笔记本电脑、平板电脑、掌上商务通等设备。计算设备130可以根据来自点云数据采集装置120的扫描结果，确定包裹外表面的点云数据(point cloud data)。点云数据是包裹外表面上点的 集合。点云数据中每个点的坐标可以表示为点云数据采集装置120的坐标系下坐标，也可以表示为世界坐标系中坐标。
上述点云数据具体为包裹哪些表面上点的集合取决于点云数据采集装置120的安装位置和角度，点云数据即为点云数据采集装置120能够采集到的包裹外表面上点的集合。例如，点云数据采集装置120位于传送带110上方，以垂直于传送带110的方向对包裹进行扫描，那么点云数据即为包裹上表面上的点以及侧面突出部分的点的集合，也就是点云数据采集装置120能够采集到的包裹外表面上点的集合。
计算设备130可以根据包裹的点云数据，自动检测包裹的规整度。这里，规整度表示包裹的表面平整程度和包裹已装载量与容量上限的接近程度。
下面结合图2对检测包裹的规整度的方式进行说明。
图2A示出了根据本申请一些实施例的检测包裹的方法200的流程图。方法200例如由计算设备130执行。
如图2A所示，在步骤S201中，获取包裹的外表面的点云数据。例如，步骤S201中计算设备可以获取包裹的上表面的点云数据。包裹的上表面是指俯视角度能够观察到的包裹的表面区域。
在一些实施例中，步骤S201中计算设备可以接收数据采集装置对传送带上包裹进行扫描而得到的扫描结果。根据扫描结果，步骤S201中计算设备可以确定包裹外表面的点云数据。点云数据中每个点可以认为对应包裹外表面上一个点。点云数据中每个点的信息包括该点的位置信息。一个点的位置信息例如可以表示为一个三维坐标系中的坐标。例如，图2B示出了根据本申请一些实施例的包裹的示意图。图2C示出了根据本申请一些实施例的点云数据的示意图。图2C中点云数据可以体现出图2B中包裹的外形特征。
在步骤S202中，根据点云数据，确定包裹的外形特征。这里，外形特征例如可以包括：顶面平整度、体积占空比、理想体积占比和轮廓矩形度中至少一个。
在步骤S203中，根据包裹的外形特征，确定包裹的规整度。其中，规整度表示包裹的表面平整程度和/或包裹已装载量与包裹容量上限的接近程度。规整度例如：为包裹的多种外形特征的加权和。例如，步骤S203中计算设备 可以计算顶面平整度、体积占空比、理想体积占比和轮廓矩形度中至少一个外形特征的加权和，并将加权和作为所述包裹的规整度。
综上，根据本申请的检测包裹的方法200，基于包裹的点云数据，可以获取包裹的外形特征，并且根据外形特征确定规整度。方法200可以避免人工检测包裹规整度的麻烦，实现由设备对包裹进行规整度的自动检测，进而提高包裹规整度检测的效率和检测结果的一致性。
在一些实施例中，步骤S202可以实施为方法300。也就是说，上述步骤S202可以包括以下步骤：
如图3所示，在步骤S301中，根据点云数据，确定包裹的顶面平整度。
其中，顶面平整度用于表征包裹顶部表面的平滑程度。顶部表面可以为俯视角度能够观察到的包裹的表面区域。
在步骤S302中，根据点云数据，确定包裹的体积占空比。
体积占空比为包裹的体积与包裹的最小外接长方体的体积之比。这里，体积占空比越高，说明包裹越接近于长方体。其中，计算设备可以根据点云数据所表示的包裹对应的各个点的坐标，进而确定包裹的体积。还可以根据点云数据所表示的包裹对应的各个点的坐标确定包裹的最小外接长方体，进而根据包裹的体积与包裹的最小外接长方体的体积之比，确定体积占空比。
在步骤S303中，根据点云数据，确定包裹的理想体积占比。
理想体积占比为包裹的体积与包裹的容量上限之比。这里，理想体积占比越高，说明包裹的装载量越靠近包裹的容量上限。其中，包裹的体积可以根据点云数据所表示的包裹对应的各个点的坐标确定。包裹的容量上限可以预先获取并存储。
在一种实施方式中，当各个包裹是容量统一的同一规格的包裹时，计算设备可以预先获取包裹的容量并进行存储，以便计算包裹的理想体积占比时使用。当各个包裹是容量不统一的不同规格的包裹时，计算设备可以预先获取各个包裹的容量，并对应于各个包裹的标识进行存储。上述点云数据采集装置120在对包裹进行扫描时，还可以通过扫描包裹上的二维码等方式获取包裹的标识，进而发送至计算设备。这样，计算设备便可以根据该标识确定其对应的包裹的容量，进而确定包裹的理想体积占比。
在步骤S304中，根据点云数据，确定包裹的轮廓矩形度。
轮廓矩形度为包裹的水平投影区域的面积与水平投影区域的最小外接矩形区域的面积之比。也就是，包裹在水平面的投影区域的面积与在水平面的投影区域的最小外接矩形区域的面积之比。轮廓矩形度越高，说明包裹的外形与长方体或立方体越接近，包裹的外形也就越规整。
在一种实施方式中，计算设备可以根据包裹的点云数据，确定包裹在水平面的投影区域，进而可以确定该投影区域的最小外接矩形，然后便可以计算包裹的水平投影区域的面积与水平投影区域的最小外接矩形区域的面积之比，确定包裹的轮廓矩形度。
综上，方法300可以从多种角度分析包裹的外形特征，从而能够利用多种外形特征确定包裹的规整度。由于外形特征可以包括顶面平整度、体积占空比、理想体积占比和/或轮廓矩形度，因此计算得到的包裹的规整度可以表示包裹的表面平整程度和/或包裹已装载量与包裹容量上限的接近程度。
图4示出了根据本申请一些实施例的检测包裹的方法400的流程图。方法400例如由计算设备130执行。
如图4所示，在步骤S401中，获取包裹的外表面的点云数据。
在步骤S402中，根据点云数据，确定包裹的外形特征。
在步骤S403中，根据包裹的外形特征，确定包裹的规整度。
由于上述步骤S401-步骤S403分别与上述步骤S201-步骤S203相同，可以参见上述步骤S201-步骤S203部分的说明，在此不再赘述。
在步骤S404中，判断规整度是否达到规整度阈值。
其中，规整度阈值可以预先根据实际需要设定，当包裹的规整度要求较高时，规整度阈值可以较高；当包裹的规整度要求较低时，规整度阈值可以较低。
在步骤S404中确定规整度达到规整度阈值时，在方法400中，继续执行步骤S405，确定包裹的分类结果为第一类别。在步骤S404中确定规整度低于规整度阈值时，在方法400中继续执行步骤S406，确定包裹的分类结果为第二类别。
其中，第一类别为满足规整度要求的类型，第二类别为不满足规整度要 求的类型。由于包裹的规整度达到规整度阈值时，说明该包裹的规整度是较高的，符合对包裹的规整度要求，所以可以确定该包裹的类别为第一类别。包裹的规整度未达到规整度阈值时，说明该包裹的规整度是较低的，不符合对包裹的规整度要求，所以可以确定该包裹的类别为第二类别。
执行步骤S406之后，可以继续执行步骤S407。在步骤S407中，将包裹的分类结果发送到分拣装置。这样，分拣装置可以根据分类结果对包裹进行分拣。例如，分拣装置可以将传送带上属于第二类别的包裹分拣到目标位置。这样，工作人员可以对第二类别的包裹可以进行整理，以便使该包裹在整理后可以满足规整度要求，从而提高包裹的运输方便性。如果包裹的类别为第一类别，说明该包裹的规整度是较高的，符合对包裹的规整度要求，所以可以进行下一步处理，例如，可以将该包裹进行装车、入库处理，或者进入下一分拣流程等，而不需要工作人员对其进行整理。
在一些实施例中，步骤S301可以实施为方法500。也就是说，上述步骤S301可以包括以下步骤：
如图5A所示，在步骤S501中，确定点云数据的主方向。主方向用于反映点云数据的整体朝向。
在一些实施例中，步骤S501可以实施为图5B所示的流程图。也就是说，步骤S501可以包括以下步骤：
如图5B所示，在步骤S5011中，计算点云数据的中心点。中心点的坐标为点云数据中点的坐标均值。
在步骤S5012中，将点云数据中各点的坐标减去中心点的坐标，得到中心偏移后的点云数据。
在步骤S5013中，确定中心偏移后的点云数据组成的第一矩阵和第二矩阵。其中，第二矩阵为第一矩阵的转置矩阵。
在步骤S5014中，计算第一矩阵和第二矩阵之积，得到协方差矩阵。例如，第一矩阵假设为A，第二矩阵为A T。协方差矩阵cov＝AA T。
在步骤S5015中，对协方差矩阵进行奇异值分解(Singular Value Decomposition，SVD)，得到协方差矩阵的特征向量，进而取特征值最大的特征向量作为点云数据的主方向。
在步骤S502中，确定点云数据的目标子集。
目标子集中每个点的法向与主方向夹角小于夹角阈值。这里，夹角阈值例如为10度、15度、18度等，但不限于此。确定点云数据的主方向之后，计算设备可以计算点云数据中每个点的法线方向与主方向之间的夹角，进而将夹角小于夹角阈值的点所组成的集合确定为点云数据的目标子集。其中，每个点的法线是通过该点垂直于该点所在处的局部平面的直线。
在步骤S503中，将目标子集的点数量与点云数据的点数量之比作为顶面平整度。
这里，顶面平整度越高，包裹顶部表面越平滑。具体来说，目标子集的点数量越多，说明对应的法向与主方向夹角小于夹角阈值的点的数量越多，那么也就说明包裹顶部的点在一个平面的数量也就越多，所以可以确定包裹顶部表面越平滑。
在一些实施例中，步骤S301可以实施为方法600。也就是说，上述步骤S301可以包括以下步骤：
如图6所示，在步骤S601中，对点云数据进行下采样操作，得到采样后的点云数据。这里，通过下采样操作，可以减小点云数据的数据规模，从而提高计算顶面平整度的效率。
在步骤S602中，基于主成分分析算法，确定采样后的点云数据的主方向。主方向用于反映点云数据的整体朝向。其中，主成分分析算法具体可以为上述奇异值分解算法，在此不再赘述。
在步骤S603中，确定采样后的点云数据的目标子集。
目标子集中每个点的法向与主方向夹角小于夹角阈值。这里，夹角阈值例如为10度，但不限于此。由于步骤S603与上述步骤S502相同，可以参见上述步骤S502部分说明，在此不再赘述。
在步骤S604中，将目标子集的点数量与采样后的点云数据的点数量之比作为顶面平整度。
例如，顶面平整度可以根据下述公式得到。
Figure PCTCN2020142175-appb-000001
其中，flat_ratio为顶面平整度，N angle为目标子集的点数量，N sampled表示采样后的点云数据的点数量。
在一些实施例中，步骤S302可以实施为方法700。也就是说，步骤S302可以包括以下步骤：
如图7所示，在步骤S701中，根据点云数据确定包裹的外形模型。换言之，步骤S701可以根据点云数据确定包裹的三维模型。由于点云数据可以表示包裹的外部的各个点的位置，因此，可以根据点云数据构建包裹的三维模型。
在步骤S702中，根据包裹的外形模型确定包裹的最小外接长方体，并确定长方体的体积。
确定了包裹的外形模型后，便可以基于该外形模型确定包裹的最小外接长方体，进而计算该最小外接长方体的体积。
在步骤S703中，根据点云数据，确定包裹的体积。这里，基于点云数据，步骤S703可以采用各种体积计算方式，确定包裹的体积。
在一些实施例中，步骤S703可以实施为方法800。也就是说，步骤S703可以包括以下步骤：
如图8所示，在步骤S801中，将点云数据投影至预定平面内，得到所述点云数据对应的投影点。该预定平面可以是包裹的承载面，例如包裹被放置在传送带上的情况，由于包裹的下表面与传送带的表面重合，所以预定平面可以为传送带的表面。
在步骤S802中，对点云数据对应的投影点进行栅格化处理，得到包含投影点的多个栅格。
在步骤S803中，计算每个栅格对应的体积，每个栅格对应的体积为该栅格的面积与投影至该栅格的点云数据的高度之积。
投影至该栅格的点云数据的高度为投影至该栅格的点云数据的高度均值或者投影至该栅格的点云数据的高度范围内、与最多数量的点对应的高度值。这里，与最多数量的点对应的高度值也可以理解为：在按照高度对投影至该栅格的点云数据进行分组后，包括的点数量最多的分组对应的高度值。点云数据中单个点的高度为该点在三维坐标系中的竖直方向的坐标值与传送带在 竖直方向的坐标值之差。这里，每个栅格对应的体积可以理解为该栅格对应的柱体的体积。柱体可以认为是栅格将包裹的外形模型竖直切分得到的柱体。
在步骤S804中，将多个栅格对应的体积之和作为包裹的体积。
其中，在步骤S804中，将多个栅格对应的体积之和作为包裹的体积也就是将所有栅格对应的体积之和作为包裹的体积。
综上，方法800通过栅格化点云数据的方式，可以计算包裹的积分体积。因此，即使包裹表面不够平整，方法800也可以较准确地确定包裹的体积。
在步骤S704中，根据长方体的体积和包裹的体积，计算体积占空比。例如，步骤S704可以根据下述公式计算体积占空比。
Figure PCTCN2020142175-appb-000002
其中，vol integral为包裹的体积，vol box为该包裹的最小外接长方体的体积。
在一些实施例中，步骤S304可以实施为方法900。也就是说，上述步骤S304可以包括以下步骤：
如图9所示，在步骤S901中，确定点云数据在水平面内的水平投影区域，并确定水平投影区域的面积。
在步骤S902中，确定水平投影区域的最小外接矩形区域，并确定矩形区域的面积。
在步骤S903中，将水平投影区域的面积与矩形区域的面积之比作为轮廓矩形度。例如，步骤S903可以根据下述方式计算轮廓矩形度。
Figure PCTCN2020142175-appb-000003
其中，rectangularity为轮廓矩形度，S convex_hull为水平投影区域的面积，S min_rect为最小外接矩形区域的面积。轮廓矩形度越接近于1，则包裹顶面越接近长方形，则包裹规整性越好。
在上述实施例中，包裹的规整度与顶面平整度、体积占空比、理想体积占比、轮廓矩形度相关，其中，顶面平整度越高，表示包裹表面的平整度越高，包裹的规整度也就越高；体积占空比和理想体积占比越高，表示包裹的剩余容量越少，包裹的规整度也就越高；轮廓矩形度越高，表示包裹的轮廓 越接近于矩形，包裹的规整度也就越高。也就是说，包裹的规整度与顶面平整度、体积占空比、理想体积占比即轮廓矩形度均呈正相关关系。
图10示出了根据本申请一些实施例的检测包裹的装置1000。装置1000例如可以部署在计算设备130中。
如图10所示，装置1000包括获取单元1001、特征提取单元1002和规整度确定单元1003。
获取单元1001，用于获取包裹的外表面的点云数据。
特征提取单元1002，用于根据点云数据，确定包裹的外形特征。
规整度确定单元1003，用于根据包裹的外形特征，确定包裹的规整度。其中，规整度表示包裹的表面平整程度和/或包裹已装载量与包裹容量上限的接近程度。装置1000更具体的实施方式与方法200一致，这里不再赘述。
综上，根据本申请的检测包裹的装置1000，基于包裹的点云数据，可以获取包裹的外形特征，并且根据外形特征确定规整度。装置1000可以避免人工检测包裹规整度的麻烦，实现由设备对包裹进行规整度的自动检测，进而提高包裹规整度检测的效率和检测结果的一致性。
图11示出了根据本申请一些实施例的检测包裹的装置1100。装置1100例如可以部署在计算设备130中。
如图11所示，装置1100包括获取单元1101、特征提取单元1102、规整度确定单元1103、分类单元1104。
获取单元1101，可以用于获取包裹的外表面的点云数据。
特征提取单元1102，可以用于根据点云数据，确定包裹的外形特征。
规整度确定单元1103，可以用于根据包裹的外形特征，确定包裹的规整度。
分类单元1104，可以用于判断规整度是否达到规整度阈值，在规整度达到规整度阈值时，分类单元1104确定包裹的分类结果为第一类别。在规整度低于规整度阈值时，分类单元1104确定包裹的分类结果为第二类别。
在一些实施例中，为了获取点云数据，获取单元1101可以接收数据采集装置对传送带上包裹进行扫描而得到的扫描结果。根据扫描结果，获取单元1101确定点云数据。
根据点云数据，特征提取单元1102可以确定包裹的一种或多种外形特征。例如，根据点云数据，特征提取单元1102确定包裹的顶面平整度。其中，顶面平整度用于表征包裹顶部表面的平滑程度。根据点云数据，特征提取单元1102确定包裹的体积占空比。体积占空比为包裹的体积与包裹的最小外接长方体的体积之比。根据点云数据，特征提取单元1102确定包裹的理想体积占比。理想体积占比为包裹的体积与包裹的容量上限之比。根据点云数据，确定包裹的轮廓矩形度。轮廓矩形度为包裹的水平投影区域的面积与水平投影区域的最小外接矩形区域的面积之比。
在一些实施例中，规整度确定单元1103，可以用于计算顶面平整度、体积占空比、理想体积占比和轮廓矩形度中至少一个外形特征的加权和，并将加权和作为包裹的规整度。
特征提取单元1102，可以用于确定点云数据的主方向。主方向用于反映点云数据的整体朝向。特征提取单元1102还可以用于确定点云数据的目标子集。目标子集中每个点的法向与主方向夹角小于夹角阈值。这样，特征提取单元1102可以将目标子集的点数量与点云数据的点数量之比作为顶面平整度。
在确定点云数据的主方向之前，特征提取单元1102还可以用于对点云数据进行下采样操作，得到采样后的点云数据。
在一些实施例中，特征提取单元1102，可以用于根据点云数据确定包裹的外形模型。根据包裹的外形模型，特征提取单元1102可以用于确定包裹的最小外接长方体，并确定长方体的体积。根据点云数据，特征提取单元1102可以确定包裹的体积。根据长方体的体积和包裹的体积，特征提取单元1102可以计算体积占空比。
为了确定包裹的体积，特征提取单元1102，可以用于按照预定平面内的栅格化阵列，对点云数据进行栅格化划分，得到多个栅格对应的点，其中，每个栅格对应的点的水平投影属于该栅格的范围。根据每个栅格对应的点，特征提取单元1102计算该栅格对应的体积。其中，该栅格对应的体积为该栅格的面积与该栅格对应的点的高度之积。特征提取单元1102将多个栅格对应的体积之和作为包裹的体积。
为了确定包裹的轮廓矩形度，特征提取单元1102，可以用于确定点云数 据在水平面内的水平投影区域，并确定水平投影区域的面积。特征提取单元1102可以确定水平投影区域的最小外接矩形区域，并确定矩形区域的面积。特征提取单元1102可以将水平投影区域的面积与矩形区域的面积之比作为轮廓矩形度。
为了确定点云数据的主方向，特征提取单元1102，可以用于计算点云数据的中心点。中心点的坐标为点云数据中点的坐标均值。特征提取单元1102将点云数据中各点的坐标减去中心点的坐标，得到中心偏移后的点云数据。特征提取单元1102确定中心偏移后的点云数据组成的第一矩阵和第二矩阵。其中第二矩阵为第一矩阵的转置矩阵。特征提取单元1102通过计算第一矩阵和第二矩阵之积，得到协方差矩阵。特征提取单元1102对协方差矩阵进行奇异值分解，取特征值最大的特征向量作为点云数据的主方向。
图12示出了根据本申请一些实施例的计算设备的示意图。如图12所示，该计算设备包括一个或者多个处理器(Central Processing Unit，CPU)1202、通信模块1204、存储器1206、用户接口1210，以及用于互联这些组件的通信总线1208。其中，图12中只示出一个处理器作为示意。
处理器1202可通过通信模块1204接收和发送数据以实现网络通信和/或本地通信。
用户接口1210包括一个或多个输出设备1212，其包括一个或多个扬声器和/或一个或多个可视化显示器。用户接口1210也包括一个或多个输入设备1214。用户接口1210例如可以接收遥控器的指令，但不限于此。其中，图12中只示出一个输出设备和一个输入设备作为示意。
存储器1206可以是高速随机存取存储器，诸如DRAM、SRAM、DDR RAM、或其他随机存取固态存储设备；或者非易失性存储器，诸如一个或多个磁盘存储设备、光盘存储设备、闪存设备，或其他非易失性固态存储设备。
存储器1206存储处理器1202可执行的指令集，包括：
操作系统1216，包括用于处理各种基本系统服务和用于执行硬件相关任务的程序；
应用1218，包括用于实现上述检测包裹的各种程序，例如可以包括检测包裹的装置1000和1100。这种程序能够实现上述各实例中的处理流程，比如 可以包括检测包裹的方法。在一种实施方式中，应用1218可以包括用于实现上述检测包裹的各种程序，例如可以包括检测包裹的装置1000或检测包裹的装置1100。
另外，本申请的每一个实施例可以通过由数据处理设备如计算机执行的数据处理程序来实现。显然，数据处理程序构成了本发明。此外，通常存储在一个存储介质中的数据处理程序通过直接将程序读取出存储介质或者通过将程序安装或复制到数据处理设备的存储设备(如硬盘和或内存)中执行。因此，这样的存储介质也构成了本发明。存储介质可以使用任何类型的记录方式，例如纸张存储介质(如纸带等)、磁存储介质(如软盘、硬盘、闪存等)、光存储介质(如CD-ROM等)、磁光存储介质(如MO等)等。
因此本申请还公开了一种非易失性存储介质，其中存储有程序。该程序包括指令，所述指令当由处理器执行时，使得计算设备执行根据本申请的检测包裹的方法。
本申请还公开了一种包含指令的计算机程序产品，当所述计算机程序产品在计算机中运行时，使得所述计算机执行上述任一实施例所述的检测包裹的方法。
根据本申请的检测包裹的方案，计算机程序产品在计算机中运行时，计算机可以基于包裹的点云数据，可以获取包裹的外形特征，并且根据外形特征确定规整度。检测包裹的方案可以避免人工检测包裹规整度的麻烦，实现由设备对包裹进行规整度的自动检测，进而提高包裹规整度检测的效率和检测结果的一致性。
另外，本申请所述的方法步骤除了可以用数据处理程序来实现，还可以由硬件来实现，例如，可以由逻辑门、开关、专用集成电路(ASIC)、可编程逻辑控制器和嵌微控制器等来实现。因此这种可以实现本申请所述方法的硬件也可以构成本申请。以上所述仅为本发明的较佳实施例，并不用以限制本发明，凡在本发明的精神和原则之内，所做的任何修改、等同替换、改进等，均应包含在本发明保护的范围之内。

Claims (15)

  1. 一种检测包裹的方法，包括：
    获取包裹的外表面的点云数据；
    根据所述点云数据，确定所述包裹的外形特征；
    根据所述包裹的外形特征，确定所述包裹的规整度，其中，所述规整度表示包裹的表面平整程度和/或包裹已装载量与包裹容量上限的接近程度。
  2. 如权利要求1所述的方法，其中，所述获取包裹的外表面的点云数据，包括：
    接收点云数据采集装置对传送带上所述包裹进行扫描而得到的扫描结果；
    根据所述扫描结果，确定所述点云数据。
  3. 如权利要求1所述的方法，其中，所述根据所述点云数据，确定所述包裹的外形特征，包括下述中至少一个步骤：
    根据所述点云数据，确定所述包裹的顶面平整度，其中，所述顶面平整度用于表征所述包裹顶部表面的平滑程度；
    根据所述点云数据，确定所述包裹的体积占空比，所述体积占空比为所述包裹的体积与所述包裹的最小外接长方体的体积之比；
    根据所述点云数据，确定所述包裹的理想体积占比，所述理想体积占比为所述包裹的体积与所述包裹的容量上限之比；
    根据所述点云数据，确定所述包裹的轮廓矩形度，所述轮廓矩形度为所述包裹的水平投影区域的面积与水平投影区域的最小外接矩形区域的面积之比。
  4. 如权利要求3所述的方法，其中，所述根据所述包裹的外形特征，确定所述包裹的规整度，包括：
    计算所述顶面平整度、体积占空比、理想体积占比和轮廓矩形度中至少一个外形特征的加权和，并将所述加权和作为所述包裹的规整度。
  5. 如权利要求3所述的方法，其中，所述根据所述点云数据，确定所述包裹的顶面平整度，包括：
    确定所述点云数据的主方向，所述主方向用于反映所述点云数据的整体朝向；
    确定所述点云数据的目标子集，目标子集中每个点的法向与所述主方向夹角小于夹角阈值；
    将所述目标子集的点数量与所述点云数据的点数量之比作为所述顶面平整度。
  6. 如权利要求5所述的方法，其中，所述确定所述点云数据的主方向，包括：
    对点云数据进行下采样操作，得到采样后的点云数据；
    确定采样后的点云数据的主方向。
  7. 如权利要求3所述的方法，其中，所述根据所述点云数据，确定所述包裹的体积占空比，包括：
    根据所述点云数据确定包裹的外形模型；
    根据所述包裹的外形模型确定所述包裹的最小外接长方体，并确定所述长方体的体积；
    根据所述点云数据，确定所述包裹的体积；
    根据所述长方体的体积和所述包裹的体积，计算所述体积占空比。
  8. 如权利要求7所述的方法，其中，所述根据所述点云数据，确定所述包裹的体积，包括：
    将所述点云数据投影至预定平面内，得到所述点云数据对应的投影点；
    对所述点云数据对应的投影点进行栅格化处理，得到包含投影点的多个栅格；
    计算每个栅格对应的体积，每个栅格对应的体积为该栅格的面积与投影至该栅格的点云数据的高度之积；
    将多个栅格对应的体积之和作为包裹的体积。
  9. 如权利要求3所述的方法，其中，所述根据所述点云数据，确定所述包裹的轮廓矩形度，包括：
    确定所述点云数据在水平面内的水平投影区域，并确定水平投影区域的面积；
    确定所述水平投影区域的最小外接矩形区域，并确定所述矩形区域的面积；
    将所述水平投影区域的面积与所述矩形区域的面积之比作为所述轮廓矩形度。
  10. 如权利要求5所述的方法，其中，所述确定所述点云数据的主方向，包括：
    计算点云数据的中心点，所述中心点的坐标为点云数据中点的坐标均值；
    将点云数据中各点的坐标减去中心点的坐标，得到中心偏移后的点云数据；
    确定中心偏移后的点云数据组成的第一矩阵和第二矩阵，其中第二矩阵为第一矩阵的转置矩阵；
    计算第一矩阵和第二矩阵之积，得到协方差矩阵；
    对协方差矩阵进行奇异值分解，取特征值最大的特征向量作为点云数据的主方向。
  11. 一种检测包裹的装置，包括：
    获取单元，用于获取包裹的外表面的点云数据；
    特征提取单元，用于根据所述点云数据，确定所述包裹的外形特征；
    规整度确定单元，用于根据所述包裹的外形特征，确定所述包裹的规整度，其中，所述规整度表示包裹的表面平整程度和包裹已装载量与容量上限的接近程度。
  12. 一种计算设备，包括：
    存储器；
    处理器；
    程序，存储在该存储器中并被配置为由所述处理器执行，所述程序包括用于执行权利要求1-10中任一项所述方法的指令。
  13. 一种存储介质，存储有程序，所述程序包括指令，所述指令当由计算设备执行时，使得所述计算设备执行如权利要求1-10中任一项所述的方法。
  14. 一种物流系统，包括：
    如权利要请求12所述的计算设备；
    传送带；
    点云数据采集装置，用于采集所述传送带上包裹的点云数据。
  15. 一种包含指令的计算机程序产品，当所述计算机程序产品在计算机中运行时，使得所述计算机执行权利要求1至10任一项所述的方法。
PCT/CN2020/142175 2019-12-31 2020-12-31 检测包裹的方法、装置、计算设备、物流系统及存储介质 Ceased WO2021136509A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
EP20909919.1A EP4086829A4 (en) 2019-12-31 2020-12-31 PACKAGING DETECTION METHOD, APPARATUS, COMPUTER DEVICE, LOGISTICS SYSTEM AND RECORDING MEDIA
KR1020227024401A KR102757317B1 (ko) 2019-12-31 2020-12-31 소포 검사 방법, 장치, 컴퓨팅 장비, 물류 시스템 및 저장 매체
JP2022539768A JP7417745B2 (ja) 2019-12-31 2020-12-31 小包を検出するための方法、装置、計算装置、物流システム及び記憶媒体

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201911413069.6A CN113065810B (zh) 2019-12-31 2019-12-31 检测包裹的方法、装置、计算设备、物流系统及存储介质
CN201911413069.6 2019-12-31

Publications (1)

Publication Number Publication Date
WO2021136509A1 true WO2021136509A1 (zh) 2021-07-08

Family

ID=76558041

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2020/142175 Ceased WO2021136509A1 (zh) 2019-12-31 2020-12-31 检测包裹的方法、装置、计算设备、物流系统及存储介质

Country Status (5)

Country Link
EP (1) EP4086829A4 (zh)
JP (1) JP7417745B2 (zh)
KR (1) KR102757317B1 (zh)
CN (1) CN113065810B (zh)
WO (1) WO2021136509A1 (zh)

Cited By (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN113947629A (zh) * 2021-10-20 2022-01-18 吉旗物联科技（上海）有限公司 一种物体体积的确定方法、装置、电子设备及存储介质
CN114688992A (zh) * 2022-04-12 2022-07-01 上海快仓智能科技有限公司 一种反光物的识别方法、装置、电子设备及存储介质
CN115069587A (zh) * 2022-06-20 2022-09-20 博睿自动化设备(广州)有限公司 一种用于芯片平面度检测的自动化设备及其使用方法
CN115825902A (zh) * 2022-01-14 2023-03-21 深圳元戎启行科技有限公司 对象的运动识别方法以及相关装置
WO2023060683A1 (zh) * 2021-10-13 2023-04-20 东南大学 一种基于三维点云模型的预制梁段表面平整度检测方法
CN116228666A (zh) * 2022-12-30 2023-06-06 梅卡曼德（北京）机器人科技有限公司 压料处理方法、装置及服务器
CN117974748A (zh) * 2022-10-24 2024-05-03 顺丰科技有限公司 包裹尺寸测量方法、装置、计算机设备及存储介质
CN118817521A (zh) * 2024-06-20 2024-10-22 中交广州航道局有限公司 一种整体铸造式斧头凿岩测试系统
CN120403431A (zh) * 2025-04-29 2025-08-01 长江武汉通信管理局 船舶通航尺度及交通流检测方法、装置、设备和介质
CN121353386A (zh) * 2025-12-17 2026-01-16 浙江华睿科技股份有限公司 目标对象的体积测量方法、电子设备及可读存储介质

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR102547130B1 (ko) * 2021-11-02 2023-06-26 주식회사 포스트큐브 우편물 부피 측정 방법
CN114155228B (zh) * 2021-12-07 2024-05-10 宁波市新铭建设工程测试有限公司 一种建材试件外形合规性快速测量方法及其装置
CN117037126B (zh) * 2023-05-11 2025-11-04 梅卡曼德(雄安)机器人科技股份有限公司 转运方法、转运设备和电子设备
US20260052274A1 (en) * 2024-08-16 2026-02-19 Interdigital Vc Holdings, Inc. Dynamic pcc with multiple reference frames

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101324275B1 (ko) * 2012-11-30 2013-11-01 (주)덕평물류 화물 볼륨 체커
CN104020095A (zh) * 2014-06-19 2014-09-03 广东中烟工业有限责任公司 一种卷烟小盒空隙率的测定方法
CN108692681A (zh) * 2017-04-05 2018-10-23 村田机械株式会社 凹陷部检测装置、输送装置以及凹陷部检测方法
CN109029253A (zh) * 2018-06-29 2018-12-18 南京阿凡达机器人科技有限公司 一种包裹体积测量方法、系统、储存介质及移动终端
CN109463969A (zh) * 2018-09-10 2019-03-15 深圳蓝胖子机器人有限公司 快递柜及其快递件的识别方法、机器人和计算机存储介质
CN109584157A (zh) * 2018-10-19 2019-04-05 苏州优纳科技有限公司 物体平面度测量方法及装置、存储介质、电子测量设备
CN109785379A (zh) * 2018-12-17 2019-05-21 中国科学院长春光学精密机械与物理研究所 一种对称物体尺寸及重量的测量方法及测量系统

Family Cites Families (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR100356016B1 (ko) * 1999-12-21 2002-10-18 한국전자통신연구원 영상인식에 의한 소포우편물 부피계측시스템 및부피계측방법
WO2015021473A1 (en) * 2013-08-09 2015-02-12 Postea, Inc. Apparatus, systems and methods for enrollment of irregular shaped objects
US20170091706A1 (en) 2015-09-25 2017-03-30 Hand Held Products, Inc. System for monitoring the condition of packages throughout transit
JP6416742B2 (ja) 2015-12-09 2018-10-31 日本電信電話株式会社 含有資源量推定装置、含有資源量推定方法、および、含有資源量推定プログラム
US10262222B2 (en) * 2016-04-13 2019-04-16 Sick Inc. Method and system for measuring dimensions of a target object
CN107388960B (zh) * 2016-05-16 2019-10-22 杭州海康机器人技术有限公司 一种确定物体体积的方法及装置
CA3125730C (en) * 2018-01-05 2023-10-24 Aquifi, Inc. Systems and methods for volumetric sizing
CN109272547A (zh) * 2018-09-17 2019-01-25 南京阿凡达机器人科技有限公司 一种包裹外接体积测量方法、系统、存储介质及移动终端

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101324275B1 (ko) * 2012-11-30 2013-11-01 (주)덕평물류 화물 볼륨 체커
CN104020095A (zh) * 2014-06-19 2014-09-03 广东中烟工业有限责任公司 一种卷烟小盒空隙率的测定方法
CN108692681A (zh) * 2017-04-05 2018-10-23 村田机械株式会社 凹陷部检测装置、输送装置以及凹陷部检测方法
CN109029253A (zh) * 2018-06-29 2018-12-18 南京阿凡达机器人科技有限公司 一种包裹体积测量方法、系统、储存介质及移动终端
CN109463969A (zh) * 2018-09-10 2019-03-15 深圳蓝胖子机器人有限公司 快递柜及其快递件的识别方法、机器人和计算机存储介质
CN109584157A (zh) * 2018-10-19 2019-04-05 苏州优纳科技有限公司 物体平面度测量方法及装置、存储介质、电子测量设备
CN109785379A (zh) * 2018-12-17 2019-05-21 中国科学院长春光学精密机械与物理研究所 一种对称物体尺寸及重量的测量方法及测量系统

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP4086829A4

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2023060683A1 (zh) * 2021-10-13 2023-04-20 东南大学 一种基于三维点云模型的预制梁段表面平整度检测方法
US12136206B2 (en) 2021-10-13 2024-11-05 Southeast University Method for detecting surface flatness of precast beam based on three-dimensional point cloud model
CN113947629A (zh) * 2021-10-20 2022-01-18 吉旗物联科技（上海）有限公司 一种物体体积的确定方法、装置、电子设备及存储介质
CN115825902A (zh) * 2022-01-14 2023-03-21 深圳元戎启行科技有限公司 对象的运动识别方法以及相关装置
CN114688992A (zh) * 2022-04-12 2022-07-01 上海快仓智能科技有限公司 一种反光物的识别方法、装置、电子设备及存储介质
CN114688992B (zh) * 2022-04-12 2024-03-12 上海快仓智能科技有限公司 一种反光物的识别方法、装置、电子设备及存储介质
CN115069587A (zh) * 2022-06-20 2022-09-20 博睿自动化设备(广州)有限公司 一种用于芯片平面度检测的自动化设备及其使用方法
CN117974748A (zh) * 2022-10-24 2024-05-03 顺丰科技有限公司 包裹尺寸测量方法、装置、计算机设备及存储介质
CN116228666A (zh) * 2022-12-30 2023-06-06 梅卡曼德（北京）机器人科技有限公司 压料处理方法、装置及服务器
CN118817521A (zh) * 2024-06-20 2024-10-22 中交广州航道局有限公司 一种整体铸造式斧头凿岩测试系统
CN120403431A (zh) * 2025-04-29 2025-08-01 长江武汉通信管理局 船舶通航尺度及交通流检测方法、装置、设备和介质
CN121353386A (zh) * 2025-12-17 2026-01-16 浙江华睿科技股份有限公司 目标对象的体积测量方法、电子设备及可读存储介质

Also Published As

Publication number Publication date
EP4086829A4 (en) 2023-01-18
KR20220114619A (ko) 2022-08-17
CN113065810B (zh) 2024-01-26
KR102757317B1 (ko) 2025-01-21
JP2023508510A (ja) 2023-03-02
EP4086829A1 (en) 2022-11-09
JP7417745B2 (ja) 2024-01-18
CN113065810A (zh) 2021-07-02

Similar Documents

Publication Publication Date Title
WO2021136509A1 (zh) 检测包裹的方法、装置、计算设备、物流系统及存储介质
CN110992356B (zh) 目标对象检测方法、装置和计算机设备
US10474524B2 (en) Point cloud filter method and apparatus
CN111210429B (zh) 点云数据划分方法及装置、障碍物检测方法及装置
CN111582054B (zh) 点云数据处理方法及装置、障碍物检测方法及装置
US11300964B2 (en) Method and system for updating occupancy map for a robotic system
US9292961B1 (en) System and method for detecting a structural opening in a three dimensional point cloud
CN110632608B (zh) 一种基于激光点云的目标检测方法和装置
US11468609B2 (en) Methods and apparatus for generating point cloud histograms
CN108648230A (zh) 一种包裹尺寸测量方法、系统、储存介质及移动终端
WO2022133770A1 (zh) 点云法向量的生成方法、装置、计算机设备和存储介质
CN113538555B (zh) 基于规则箱体的体积测量方法、系统、设备及存储介质
CN115321090A (zh) 机场行李自动接取方法、装置、设备、系统及介质
CN113379826A (zh) 物流件的体积测量方法以及装置
WO2023005195A1 (zh) 地图数据的处理方法、装置、家用电器和可读存储介质
CN118258329A (zh) 基于三维扫描技术的墙面垂直度及平整度检测方法
CN110517354B (zh) 用于三维点云分割的方法、装置、系统及介质
Kaushik et al. Accelerated patch-based planar clustering of noisy range images in indoor environments for robot mapping
CN116645408A (zh) 一种基于点云数据的体积测量方法及装置
US20220128347A1 (en) System and method to measure object dimension using stereo vision
CN113538554A (zh) 基于深度相机的体积测量方法、系统、设备及存储介质
WO2025096395A1 (en) Image-assisted region growing for object segmentation and dimensioning
WO2024061101A1 (zh) 非常规件识别方法、装置、计算机设备及存储介质
US11321914B1 (en) System for generating a navigational map of an environment
Donné et al. Point Triangulation Through Polyhedron Collapse Using the l [?] Norm

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: 20909919

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2022539768

Country of ref document: JP

Kind code of ref document: A

ENP Entry into the national phase

Ref document number: 20227024401

Country of ref document: KR

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2020909919

Country of ref document: EP

Effective date: 20220801