WO2019062818A1 - 用于识别电气设备的状态的方法和装置 - Google Patents

用于识别电气设备的状态的方法和装置 Download PDF

Info

Publication number
WO2019062818A1
WO2019062818A1 PCT/CN2018/108037 CN2018108037W WO2019062818A1 WO 2019062818 A1 WO2019062818 A1 WO 2019062818A1 CN 2018108037 W CN2018108037 W CN 2018108037W WO 2019062818 A1 WO2019062818 A1 WO 2019062818A1
Authority
WO
WIPO (PCT)
Prior art keywords
electrical device
state
image
feature
predefined
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/CN2018/108037
Other languages
English (en)
French (fr)
Inventor
费维和
德雷纳伊曼纽尔
皮洛弗雷德里克
劳林皮埃尔
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.)
Schneider Electric Industries SAS
Original Assignee
Schneider Electric Industries SAS
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 Schneider Electric Industries SAS filed Critical Schneider Electric Industries SAS
Priority to EP18860822.8A priority Critical patent/EP3678048A4/en
Priority to US16/652,102 priority patent/US11302099B2/en
Publication of WO2019062818A1 publication Critical patent/WO2019062818A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Program-control systems
    • G05B19/02—Program-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B19/00—Program-control systems
    • G05B19/02—Program-control systems electric
    • G05B19/418—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM]
    • G05B19/4183—Total factory control, i.e. centrally controlling a plurality of machines, e.g. direct or distributed numerical control [DNC], flexible manufacturing systems [FMS], integrated manufacturing systems [IMS] or computer integrated manufacturing [CIM] characterised by data acquisition, e.g. workpiece identification
    • 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
    • G06V10/24—Aligning, centring, orientation detection or correction of the image
    • G06V10/243—Aligning, centring, orientation detection or correction of the image by compensating for image skew or non-uniform image deformations
    • 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
    • G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/20—Scenes; Scene-specific elements in augmented reality scenes
    • 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
    • G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
    • G06V20/63—Scene text, e.g. street names
    • G—PHYSICS
    • G05—CONTROLLING; REGULATING
    • G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
    • G05B2219/00—Program-control systems
    • G05B2219/30—Nc systems
    • G05B2219/32—Operator till task planning
    • G05B2219/32128—Gui graphical user interface
    • 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
    • G06V10/24—Aligning, centring, orientation detection or correction of the image
    • G06V10/247—Aligning, centring, orientation detection or correction of the image by affine transforms, e.g. correction due to perspective effects; Quadrilaterals, e.g. trapezoids
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/02—Recognising information on displays, dials, clocks
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V2201/00—Indexing scheme relating to image or video recognition or understanding
    • G06V2201/06—Recognition of objects for industrial automation

Definitions

  • Embodiments of the present disclosure relate to the electrical field and, more particularly, to methods and apparatus for identifying a state of an electrical device.
  • the digital monitoring of electrical equipment is an essential element for the digital management and network management of industrial equipment.
  • the usual electrical devices are designed with components that are convenient for the user to observe, such as the expansion and contraction of the button, the brightness or darkness of the indicator light, the change of the position of the handle of the circuit breaker, and the indication of the instrument pointer. Value changes and so on.
  • These visual components are easily identifiable to the human eye, but in the era of automated information, we hope to remotely detect the state of electrical equipment through electronic methods, and intelligently monitor and analyze the information through digital information.
  • the general method of camera image recognition is to first preprocess the image, including color extraction, grayscale conversion, noise reduction processing, contrast enhancement and other processing techniques, and then scan the entire preprocessed image to determine whether there is electrical equipment.
  • State-related target graphic features through known image recognition methods, such as spatial matching algorithm, target edge extraction algorithm, color extraction and matching algorithms to analyze the degree of agreement between the image and the electrical device target features, in order to judge the electrical equipment status.
  • the disadvantage of this method is that the image preprocessing calculation is large, and the preprocessing parameters need to be adjusted according to the actual lighting environment. Because the whole image needs to be scanned and judged, the calculation amount is large, the efficiency is not high, and in practical application. Due to the variety of electrical equipment models, the variability of ambient lighting, and the uncertainty of camera position, the stability of target recognition is poor, and the versatility is not strong, which is likely to cause misjudgment.
  • Embodiments of the present disclosure are directed to a method, apparatus, and system that at least partially address the above-discussed problems of the prior art.
  • a method for identifying a state of an electrical device includes: acquiring an image of the electrical device in a live environment, wherein the electrical device is disposed on a plane of a predefined polygon in the live environment; by restoring the predefined polygon in the image to Deriving the original appearance of the predefined polygon to obtain an original image of the electrical device in the field environment; and determining a state of the electrical device based on an original image of the electrical device in the live environment.
  • the method further includes receiving a selection of the electrical device in the original image to determine a location of the electrical device in the original image and as a predefined of the electrical device Positioning, and determining a first feature image associated with the electrical device in a first state; responsive to a state of the electrical device changing from the first state to the second state, acquiring a second original image and determining A second feature image associated with the electrical device in a second state.
  • the state of the electrical device has a digital feature
  • determining the state of the electrical device includes: acquiring a predefined location and feature image of the electrical device and a state corresponding to the feature image; In the original image, searching for an area that matches the feature image based on the predefined location; and determining a state of the electrical device based on a state corresponding to the feature image.
  • the state of the electrical device has an analog quantity feature
  • determining the state of the electrical device includes: acquiring a predefined location of the electrical device and a plurality of feature images and separately from the plurality of feature images Corresponding plurality of states, the plurality of states corresponding to a plurality of values of the analog feature of the electrical device; within the original image, searching for regions matching the plurality of feature images based on the predefined location Identifying an indication of a state of the analog feature in the region; and a predefined relationship between the change in state of the electrical device and the plurality of values based on an indication of a state of the analog feature To determine the status of the electrical device.
  • an apparatus for identifying a state of an electrical device includes an image capture device configured to acquire an image of the electrical device in a live environment, wherein the electrical device is disposed on a plane of a predefined polygon in the field environment; the processing device configured to: Obtaining an original image of the electrical device in the field environment by restoring the predefined polygon in the image to an original appearance of the predefined polygon; and based on the electrical device in the field environment The original image determines the state of the electrical device.
  • an apparatus for identifying a status of an electrical device includes: a processing unit; and a memory coupled to the processing unit and including instructions stored thereon, the instructions, when executed by the processing unit, causing the device to perform an action, the action comprising: receiving An image of an electrical device in a field environment, wherein the electrical device is disposed on a plane of a predefined polygon in the field environment; by restoring the predefined polygon in the image to the predefined polygon Obtaining an original image of the electrical device in the field environment; and determining a state of the electrical device based on an original image of the electrical device in the field environment.
  • a computer readable storage medium storing instructions that, in response to execution by a computing device, cause the computing device to perform an action, the action comprising: receiving an electrical device in a live environment An image in which the electrical device is disposed on a plane of a predefined polygon in the field environment; the predefined polygon in the image is restored to an original appearance of the predefined polygon to obtain the electrical device An original image in the field environment; and determining a state of the electrical device based on an original image of the electrical device in a field environment.
  • Embodiments of the present disclosure can solve the problem that the camera visual recognition speed is slow and the misjudgment is recognized, and the recognition performance is susceptible to being affected by the scene environment and camera position changes.
  • fast and stable camera visual recognition can be achieved to digitize the electrical device status visual recognition to a practical stage.
  • the camera image recognition processing is small in calculation amount and high in recognition efficiency.
  • the target recognition performance is stable, with higher accuracy and reliability.
  • the user can re-calculate the computer on-site recognition target, reconfigure the identification content without changing the source program. In this way, it can adapt to the various models and layouts of various field devices to adapt to different lighting environments.
  • these embodiments can also have a function of automatically adapting to changes in camera position and angle, and are more tolerant to camera position changes.
  • FIG. 1 shows a block diagram of an electrical device state recognition system in accordance with one embodiment of the present disclosure
  • Figure 2 shows a schematic diagram of the distortion of the target in the camera image
  • Figure 3 shows a schematic diagram of how to perform perspective correction on a distorted image
  • FIG. 4 shows a schematic diagram of a visualization rectangle in accordance with one embodiment of the present disclosure
  • FIG. 5 shows a schematic diagram of a visualization rectangle in accordance with another embodiment of the present disclosure
  • FIG. 6 shows a flow chart of a state recognition method for an electrical device, in accordance with one embodiment of the present disclosure
  • FIG. 7 illustrates a block diagram of an apparatus for electrical device state recognition, in accordance with one embodiment of the present disclosure
  • FIG. 8 shows a flow chart of a method for electrical device state identification, in accordance with one embodiment of the present disclosure.
  • FIG. 1 illustrates a block diagram of a system 100 for state recognition of an electrical device, in accordance with one embodiment of the present disclosure.
  • system 100 includes an electrical equipment field 120, an image acquisition device 140, and a processing device 160.
  • the electrical equipment site 120 is the environment in which the electrical equipment is located, and includes various electrical equipment including, but not limited to, electrical equipment or components such as circuit breakers, digital display meters, pointer meters, indicator lights, buttons, and the like.
  • electrical equipment or components such as circuit breakers, digital display meters, pointer meters, indicator lights, buttons, and the like.
  • the mounting location of these electrical devices or components is fixed, ie, the position is not easily changed once the installation is complete.
  • Embodiments of the present disclosure may utilize this feature to identify the status of an electrical device.
  • the image capture device 140 may be a device for capturing images, such as a camera, a video camera, or the like. For convenience of description, the image capturing device is sometimes simply referred to as a camera hereinafter.
  • the circuit breaker, digital display meter, pointer meter, indicator light, button, etc. in the image obtained by the image acquisition device 140 are objects that the system 100 or the processing device 160 needs to recognize.
  • Processing device 160 can be, for example, a computer. Due to the difference in position and shooting angle of the image acquisition device 140, the image formed by the field electrical device on the two-dimensional projection plane of the camera has geometric distortion.
  • some embodiments of the present disclosure arrange a reference having a predefined polygonal shape at the electrical device site 120. For example, a rectangular frame can be placed at the electrical equipment site 120 and the electrical equipment placed on the plane of the rectangular frame. Preferably, the electrical device can be placed within a rectangular frame.
  • Figure 2 shows a schematic diagram of the distortion of the target in the camera image.
  • the rectangle 210 may be trapezoidal 220, or other irregular quadrilateral, in the image taken by the camera.
  • This geometric distortion can theoretically be described by a perspective transformation formula. Let the coordinates of a point be (u, v), and the coordinates of the image become (x, y) after being projected by the camera. Taking into account the translation, scaling, rotation and point projection of the coordinate system, its coordinate transformation theoretically satisfies the following perspective transformation formula
  • FIG. 3 shows a schematic diagram of how the distortion correction image is perspective corrected using a perspective calibration method in which the trapezoid 220 is calibrated to a rectangle 210.
  • the key to perspective calibration is how to obtain a perspective transformation matrix.
  • a method of placing known geometric polygons in the field to determine a perspective transformation matrix is presented.
  • geometric polygons of known shape are placed in the field environment.
  • the polygon may be a known rectangle, and the target to be identified may be placed on the plane of the rectangle, such as within the range of known rectangles.
  • the coordinates of the four vertices of the rectangle are known, and are set to ⁇ (u1, v1), (u2, v1), (u1, v2), (u2, v2) ⁇ .
  • the perspective transformation matrix [a ij ] can be calculated by the above perspective transformation formula. Since the target to be recognized is on the plane of the known rectangle, each pixel of the camera image corresponding to the rectangle can be inversely transformed in perspective, so that the original image of the recognition target can be obtained.
  • the processing device 160 detects that the position of the vertex of the known geometric polygon has changed in the camera image.
  • the system 100 recalculates the perspective transformation matrix according to the new vertex position coordinates according to the above method, and then performs perspective calibration on the camera image according to the new perspective transformation matrix, thereby obtaining the original image of the recognition target.
  • This process can also be referred to as a perspective calibration adjustment mode. Since this process can be dynamic in real time, the system can always obtain the original image of the recognition target. This provides a reliable guarantee for subsequent image recognition.
  • FIG. 4 shows a schematic diagram of a visualization rectangle in accordance with one embodiment of the present disclosure.
  • a visual rectangular frame can be placed at the edge of the electrical equipment enclosure to facilitate identification and tracking of the rectangular frame by the computer.
  • visualization features can be indicated by a specific color.
  • FIG. 5 shows a schematic diagram of a visualization rectangle in accordance with another embodiment of the present disclosure.
  • four visualized labels can be placed on the edge of the electrical equipment enclosure such that the four visualized labels form a rectangle as a vertex.
  • visualization features can be indicated by a specific color.
  • the location of the electrical device can be more conveniently defined to facilitate identification of the state of the electrical device.
  • Figures 4 and 5 illustrate two embodiments of providing a visual rectangle at the edge of an electrical equipment enclosure
  • the predefined polygons may also be placed at other locations in the electrical equipment enclosure or other environment as long as the electrical equipment is Predetermine the conditions in the plane of the polygon.
  • the state of the electrical device can be quickly identified using computer field learning.
  • the computer field learning method is to establish a target recognition configuration database in advance before the system starts image recognition. For example, in the original image of the field device after calibration, through the man-machine operation, the image regions including the state features of the electrical device are selected one by one, and the position information of these regions and all the image images reflecting the state change of the electrical device are all saved to form a target. Identify the configuration database.
  • the system reads the target recognition configuration database, searches and matches the specified electrical device feature image according to the image position specified by the database, and determines the electrical device according to the matching result.
  • Digital features such as opening, closing, splitting, closing, etc.
  • the analog position image information predefined by the configuration database is identified in the local range (for example, an image corresponding to the maximum and minimum values of the analog quantity) Information), combined with the image analysis algorithm to obtain the analog value of the electrical equipment.
  • the recognition process since the system does not need to scan and discriminate the entire image range, only the specified range is identified and analyzed, so that rapid discrimination can be realized.
  • the feature images of each device state are separately defined, the target recognition selectivity is strong, and the false positive rate is extremely low.
  • FIG. 6 illustrates a flow chart of a method 600 of identifying a state of an electrical device, in accordance with one embodiment of the present disclosure.
  • FIG. 6 may be implemented by system 100 shown in FIG. 1, and in particular, by image acquisition device 140 and processing device 160 in system 100.
  • image acquisition device 140 obtains an image of the environment in which the electrical device is located, wherein the electrical device is disposed on a plane of a predefined polygon in the environment.
  • This environment may also be referred to as an electrical equipment site or a field environment, such as the electrical equipment site 120 shown in FIG.
  • the predefined polygon is a rectangle and the coordinates of the vertices of the rectangle are known.
  • the environment includes a visualization frame having a polygonal shape, for example, as shown in FIG.
  • the environment includes a visual label disposed at a vertex of the polygon, for example, as shown in FIG.
  • processing device 160 obtains an original image of the electrical device in the live environment by restoring the predefined polygon in the image to the original appearance of the predefined polygon.
  • the original image can be obtained by the perspective calibration method described in conjunction with FIGS. 2 to 3.
  • processing device 160 determines the status of the electrical device based on the original image of the environment. This can be done by computer field learning.
  • the method 600 further includes receiving a selection of the electrical device in the original image to determine the location of the electrical device in the original image and using the location as a predefined location for the electrical device. Additionally, a first feature image associated with the electrical device in the first state can also be determined. This feature image may be part of the original image at a predefined location of the electrical device.
  • the second original image can be acquired and the second feature image associated with the electrical device in the second state can be determined.
  • the change of state can be achieved by automatic control or manual operation, and the disclosure is not limited herein.
  • the first state can be an open state and the second state can be an off state. In this case, corresponding feature images are acquired for the open state and the closed state, respectively.
  • the state of the electrical device has an analog quantity feature, the first state and the second state corresponding to a first value and a second value of an analog feature of the electrical device, and
  • the change in state of the electrical device has a predefined relationship with the first and second values.
  • the first value and the second value can be the maximum and minimum values of the analog quantity.
  • the change in the analog feature can have a predefined linear or exponential relationship with the two values, and the like.
  • different colors can be used to indicate different temperatures.
  • the first value and the second value may correspond to two different colors or temperatures.
  • the change in color or temperature can have a predefined relationship with two values, for example in the form of a table.
  • the learning mode is the basis of the normal working mode.
  • Several embodiments of the normal operating mode will be described below in connection with analog and digital quantities, respectively.
  • the state of the electrical device has a digital feature
  • determining the state of the electrical device includes: obtaining a predefined location and feature image of the electrical device and a state corresponding to the feature image; searching within the original image based on the predefined location An area that matches the feature image; and determines a state of the electrical device based on a state corresponding to the feature image.
  • regions that match the feature image can be searched within a certain range around the predefined location. In this way, the entire image can be processed without determining the state of the electrical device, and only the image portion most relevant to the electrical device can be processed, thereby saving computational resources and improving computational efficiency.
  • the state of the electrical device has an analog quantity feature
  • determining the state of the electrical device includes: acquiring a predefined location of the electrical device and a plurality of feature images and a plurality of corresponding to the plurality of feature images respectively a state, the plurality of states corresponding to a plurality of values of an analog feature of the electrical device; within the original image, searching for an area that matches the plurality of feature images based on a predefined location; Identifying an indication of a state of the analog feature; and determining, based on an indication of a state of the analog feature, the electrical device by a predefined relationship between a change in state of the electrical device and the plurality of values status.
  • multiple states may correspond to the maximum and minimum values of the analog quantities. For the sake of clarity, it will not be repeated here.
  • FIG. 7 shows a block diagram of an electrical device state identification device 700 in accordance with one embodiment of the present disclosure.
  • the apparatus 700 includes a camera module 760, a main control module 740, an image recognition module 710, a target recognition configuration database 720, a human machine interface module 770, an electrical identification status module 730, a communication module 750, and a power supply module 780.
  • the above functional modules are divided only to facilitate the description of the image recognition method and principle. In a specific application, multiple modules may be combined in one module, one module may be further split into multiple modules, and one or more of the modules may also be omitted.
  • the camera module 760 can acquire an image of the electrical device site and transmit the image data to the main control module 740.
  • the main control module 740 transmits the image acquired by the camera module 760 to the image recognition module 710 for image recognition.
  • the image recognition module 710 receives the image data transmitted by the main control module 740, performs image calibration and image recognition in conjunction with the predefined information of the target recognition configuration database 720, and transmits the digitized recognition result to the electrical device status module 730.
  • the electrical equipment status module 730 maintains status information for the field electrical equipment and, in some embodiments, digitizes the status information of the electrical equipment for ease of management and transmission.
  • the information in the target recognition configuration database 740 is obtained by the user using the human-machine interface module 770 to perform on-the-spot computer learning of the target image to be recognized by human-machine operation.
  • the main control module 740 reads the data in the electrical device status module 730 and processes the data.
  • the communication module 750 transmits electrical device status information to the remote server for digital, networked information sharing and decision analysis.
  • the power module 780 can provide power to the modules in the device 700.
  • FIG. 7 shows a flowchart of a method 800 of identifying the status of an electrical device, in accordance with one embodiment of the present disclosure.
  • the master module 740 retrieves an image of the electrical device site from the camera module 760.
  • the electrical equipment site includes one or more electrical devices or components and is provided with a polygon having a predefined shape.
  • the master module 740 identifies the coordinates of the vertices of the polygon in the image and fluoroscopy the image to obtain the original image of the field device.
  • the original image can be obtained by the perspective calibration method described in conjunction with FIGS. 2 to 3.
  • the main control module 740 acquires the current working mode of the system from the human machine interface module 770, wherein the working mode may include: a perspective calibration adjustment mode, a computer live learning mode, a normal working mode, and the like.
  • the coordinate definition of the polygon vertices is acquired by the human interface module 770 at 810, the perspective calibration is initiated, and the perspective calibration parameters (eg, the perspective transformation matrix as described above) are saved at 812.
  • the target is identified in the configuration database 720. The system can then proceed to the next loop and return to 702.
  • the location and area of the electrical device feature image are selected in the field device original image by the human interface module at 820, and the relationship between each device state and the feature image is defined, and the target is established.
  • the configuration database 720 is identified. For example, if the electrical device is a switch having two states of on and off, the original images of the scene are respectively acquired for the two states of on and off, and the region of the feature image corresponding to the switch and the image of the feature are respectively selected. The corresponding state is also called the target recognition parameter. In this manner, the target recognition configuration database 720 is established. The system can then proceed to the next loop and return to 702.
  • the original image of the scene can be obtained for the limit state within a predefined range and the corresponding analog value can be defined.
  • the original image of the scene can be obtained for the start position and the end position in the meter pointer, and the corresponding analog value is defined for the two images.
  • image recognition module 710 reads the target identification parameters in target identification configuration database 720.
  • an image matching the specified target may be searched for in the original image according to the region specified by the target recognition parameter.
  • the device site is usually kept fixed, so that the search can be performed in a designated area without searching the entire image, thereby achieving higher accuracy and saving computation.
  • state information for the electrical device is determined 836.
  • electrical devices eg, switches
  • the state of the electrical device can be determined directly as the images that match in the target identification database 720 have corresponding states.
  • the feature image of the limit state of the analog quantity defined in the configuration database 720 is identified, and the analog value of the device is obtained in combination with the image analysis algorithm.
  • image recognition module 710 can update the information in electrical device status module 730 based on the recognition results.
  • the error information is saved in electrical device status module 730.
  • the master module 740 can transmit information in the electrical device status module 730 to the remote network via the communication module 760 at 840. The system can then proceed to the next loop and return to 702.
  • the camera image recognition processing is small in calculation amount and high in recognition efficiency.
  • the target recognition performance is stable, with higher accuracy and reliability. It is convenient for the user to re-run the computer on-site learning target and reconfigure the identification content without changing the source program. Therefore, the method can adapt to various types of field devices and layouts, and adapt to various lighting environments.
  • the method has a function of automatically adapting to changes in the position and angle of the camera, and has a large tolerance to changes in the position of the camera.
  • FIGS. 1-8 Methods, apparatus, and systems for identifying states of electrical devices in accordance with some embodiments of the present disclosure are described above in conjunction with FIGS. 1-8. It should be understood that the functions described above herein may be performed at least in part by one or more hardware logic components.
  • exemplary types of hardware logic components include: Field Programmable Gate Array (FPGA), Application Specific Integrated Circuit (ASIC), Application Specific Standard Product (ASSP), System on Chip (SOC), Complex Programmable Logical devices (CPLD) and more.
  • Program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages.
  • the program code may be provided to a general purpose computer, a special purpose computer or a processor or controller of other programmable data processing apparatus such that the program code, when executed by the processor or controller, causes the functions specified in the flowcharts and/or block diagrams/ The operation is implemented.
  • the program code may execute entirely on the machine, partly on the machine, as part of the stand-alone software package, and partly on the remote machine or entirely on the remote machine or server.
  • a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device.
  • the machine readable medium can be a machine readable signal medium or a machine readable storage medium.
  • a machine-readable medium can include, but is not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any suitable combination of the foregoing.
  • machine readable storage media may include electrical connections based on one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read only memory (ROM), an erasable programmable read only memory (EPROM or flash memory), optical fiber, compact compact disk read only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the foregoing.
  • RAM random access memory
  • ROM read only memory
  • EPROM or flash memory erasable programmable read only memory
  • CD-ROM compact compact disk read only memory
  • magnetic storage device or any suitable combination of the foregoing.

Landscapes

  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • General Engineering & Computer Science (AREA)
  • Automation & Control Theory (AREA)
  • Quality & Reliability (AREA)
  • Manufacturing & Machinery (AREA)
  • Computing Systems (AREA)
  • General Health & Medical Sciences (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Evolutionary Computation (AREA)
  • Databases & Information Systems (AREA)
  • Artificial Intelligence (AREA)
  • Health & Medical Sciences (AREA)
  • Image Analysis (AREA)

Abstract

本公开的实施例涉及用于自动识别电气设备的状态的方法和装置。该方法包括：获取所述电气设备在现场环境中的图像，其中所述电气设备设置在所述现场环境中的预定义多边形的平面上；通过将所述图像中的所述预定义多边形还原成所述预定义多边形的原貌来获得所述电气设备在现场环境中的原貌图像；以及基于所述电气设备在现场环境中的原貌图像，确定所述电气设备的状态。

Description

用于识别电气设备的状态的方法和装置 技术领域
本公开的实施例涉及电气领域，并且更具体地涉及用于识别电气设备的状态的方法和装置。
背景技术
电气设备的数字化监控是实现工业设备数字化管理和网络化管理的基本要素。为了便于人们对设备状态的掌控，通常的电气设备都设计了方便用户观察的其状态的部件，例如按钮的伸缩、指示灯的亮暗或颜色的变化、断路器手柄位置的改变、仪表指针指示值的变化等等。这些可视化部件对于人眼来讲很容易能识别出来，但是在自动化信息时代，我们更希望能通过电子的方法远程探测到电气设备的状态，并通过信息数字化进行智能监控和决策分析。
目前已经出现了利用摄像机自动识别电气设备状态的相关方案，但是受限于现场环境的复杂多变性和摄像机自动识别性能不够稳定等因素，这一技术还没有达到实用的阶段。
一般的摄像机图像识别的方法是，首先对图像进行预处理，包括颜色提取、灰阶转换、降噪处理、对比度增强等处理技术，然后扫描整幅预处理后的图像来判断是否存在与电气设备状态相关的目标图形特征，通过已知的图像识别方法，如空间匹配算法、目标边缘提取算法、颜色提取和匹配算法等工具来分析图像与电气设备目标特征的吻合程度，以此来判断电气设备的状态。这种方法的缺点是图像预处理计算量大，预处理参数需要根据实际的照明环境的不同而调整，由于需要对整幅图像进行扫描判断，计算量大，效率不高，而且在实际应用中由于受电气设备型号的多样性、环境照明的多变性、摄像机位置的不确定性等影响，造成目标识别的稳定性差，通用性不强，容易造成误判。
发明内容
本公开的实施例目的在于提供至少部分地解决现有技术的上述问题的方法、设备和系统。
根据一些实施例，提供了一种用于识别电气设备的状态的方法。该方法包括：获取所述电气设备在现场环境中的图像，其中所述电气设备设置在所述现场环境中的预定义多边形的平面上；通过将所述图像中的所述预定义多边形还原成所述预定义多边形的原貌来获得所述电气设备在所述现场环境中的原貌图像；以及基于所述电气设备在所述现场环境中的原貌图像，确定所述电气设备的状态。
在一些实施例中，该方法还包括：接收在所述原貌图像中对所述电气设备的选择，以确定所述电气设备在所述原貌图像中的位置，并作为所述电气设备的预定义位置，并且确定与所述电气设备处于第一状态相关联的第一特征图像；响应于所述电气设备的状态从所述第一状态改变到所述第二状态，获取第二原貌图像并确定与所述电气设备处于第二状态相关联的第二特征图像。
在一些实施例中，所述电气设备的状态具有数字量特征，确定所述电气设备的状态包括：获取所述电气设备的预定义位置和特征图像以及与所述特征图像对应的状态；在所述原貌图像内，基于所述预定义位置搜索与所述特征图像匹配的区域；以及基于与所述特征图像对应的状态来确定所述电气设备的状态。
在一些实施例中，所述电气设备的状态具有模拟量特征，并且确定所述电气设备的状态包括：获取所述电气设备的预定义位置和多个特征图像以及与所述多个特征图像分别对应的多个状态，所述多个状态对应于所述电气设备的模拟量特征的多个数值；在所述原貌图像内，基于所述预定义位置搜索与所述多个特征图像匹配的区域；在所述区域中识别所述模拟量特征的状态的指示；以及基于所述模拟量特征的状态的指示，通过所述电气设备的状态的变化与所述多个数值之间的预定义关系来确定所述电气设备的状态。
根据一些实施例，提供了一种用于识别电气设备的状态的装置。该装置包括：图像采集设备，被配置为获取所述电气设备在现场环境中的图像，其中所述电气设备设置在所述现场环境中的预定义多边形的平面上；处理设备，被配置为：通过将所述图像中的所述预定义多边形还原成所述预定义多边形的原貌来获得所述电气设备在所述现场环境中的原貌图像；以及基于所述电气设备在所述现场环境中的原貌图像，确定所述电气设备的状态。
根据一些实施例，提供了一种用于识别电气设备的状态的设备。该设备包括：处理单元；以及存储器，耦合至所述处理单元并且包括存储于其上的指令，所述指令在由所述处理单元执行时使所述设备执行动作，所述动作包括：接收所述电气设备在现场环境中的图像，其中所述电气设备设置在所述现场环境中的预定义多边形的平面上；通过将所述图像中的所述预定义多边形还原成所述预定义多边形的原貌来获得所述电气设备在所述现场环境中的原貌图像；以及基于所述电气设备在所述现场环境中的原貌图像，确定所述电气设备的状态。
根据一些实施例，提供了一种存储有指令的计算机可读存储介质，所述指令响应于由计算设备的执行而使得所述计算设备执行动作，所述动作包括：接收电气设备在现场环境中的图像，其中所述电气设备设置在所述现场环境中的预定义多边形的平面上；将所述图像中的所述预定义多边形还原成所述预定义多边形的原貌来获得所述电气设备在所述现场环境中的原貌图像；以及基于所述电气设备在现场环境中的原貌图像，确定所述电气设备的状态。
本公开的实施例可以解决相机视觉识别速度慢并且识别误判、以及识别性能容易受现场环境和相机位置改变而受影响的问题。在一些实施例中，可以实现快速稳定的相机视觉识别，使电气设备状态视觉识别数字化达到实用的阶段。在这些实施例中，相机图像识别处理计算量小，识别效率高。此外，目标识别的性能稳定，具备更高的正确率和可靠性。而且，用户可以对现场识别目标重新进行计算机现场学习，重新配置识别内容而不需要改变源程序。以这种方式，可以适应 各种现场设备的型号和布局，适应各种不同的照明环境。而且，这些实施例还可以具备自动适应相机位置和角度变化的功能，对相机位置变化的宽容性大。
附图说明
图1示出了根据本公开的一个实施例的电气设备状态识别系统的框图；
图2示出了目标在相机图像中的畸变的示意图；
图3示出了如何对畸变图像进行透视校正的示意图；
图4示出了根据本公开的一个实施例的可视化矩形的示意图；
图5示出了根据本公开的另一实施例的可视化矩形的示意图；
图6示出了根据本公开的一个实施例的用于电气设备的状态识别方法的流程图；
图7示出了根据本公开的一个实施例的用于电气设备状态识别的装置的框图；以及
图8示出了根据本公开的一个实施例的用于电气设备状态识别的方法的流程图。
具体实施方式
图1示出了根据本公开的一个实施例的用于电气设备的状态识别的系统100的框图。如图1所示，系统100包括电气设备现场120、图像采集设备140和处理设备160。电气设备现场120是电气设备所在的环境，其包括各种电气设备，包括但不限于：断路器、数码显示仪表、指针式仪表、指示灯、按钮等电气设备或部件。一般而言，这些电气设备或部件的安装位置是固定的，即，一旦安装完成不会轻易改变其位置。本公开的实施例可以利用这一特性来对电气设备的状态进行识别。
图像采集设备140可以是照相机、摄像机等用于采集图像的设备。为描述方便起见，下文有时将图像采集设备简称为相机。图像采集设 备140所获得图像中的断路器、数码显示仪表、指针式仪表、指示灯、按钮等是系统100或处理设备160需要识别的对象。处理设备160可以是例如计算机。由于图像采集设备140所处的位置和拍摄角度的不同，现场电气设备在相机二维投影平面上所成的像存在几何畸变。为了识别这些电气设备或部件的状态，本公开的一些实施例在电气设备现场120布置了具有预定义多边形形状的参照物。例如，可以在电气设备现场120放置矩形框架，并且将电气设备放置在矩形框架所在平面上。优选地，可以将电气设备设置在矩形框架内。
图2示出了目标在相机图像中的畸变的示意图。矩形210在相机拍摄的图像中可能是梯形220，或者其他不规则的四边形。这种几何畸变在理论上可以用透视变换公式来加以描述。设一个点的坐标为(u，v)，经相机投影成像后，其坐标变为(x，y)。考虑到坐标系统的平移、缩放、旋转和点投影，其坐标变换在理论上满足以下透视变换公式
Figure PCTCN2018108037-appb-000001
Figure PCTCN2018108037-appb-000002
其中矩阵
Figure PCTCN2018108037-appb-000003
为透视变换矩阵。
如果知道了透视变换矩阵，就可以针对目标的几何位置坐标{(u1，v1)，(u2，v2)，…(un，vn)}来计算目标在相机图像中的坐标位置{(x1，y1)，(x2，y2)，…(xn，yn)}。反过来，也可以针对目标在相机图像中的位置通过透视逆变换来算得目标的原貌坐标位置。这种方法就称为透视校准法。图3示出了如何使用透视校准法对畸变图像进行透视校正的示意图，其中将梯形220校准为矩形210。
透视校准法的关键是怎样获得透视变换矩阵。根据本公开的一些实施例，提出了一种在现场放置已知几何多边形的方法来确定透视变换矩阵。具体而言，在现场环境中放置具有已知形状的几何多边形。例如，该多边形可以是已知的矩形，并且可以将需要识别的目标设置 在这个矩形所在的平面上，例如在已知矩形的范围内。矩形四个顶点坐标已知，设为{(u1，v1)，(u2，v1)，(u1，v2)，(u2，v2)}。从相机所拍摄的图像中，可以获取这四个顶点的坐标{(x1，y1)，(x2，y2)，(x3，y3)，(x4，y4)}。因此，可以通过以上透视变换公式计算出透视变换矩阵[a ij]。由于所要识别的目标在这个已知矩形的平面上，则可以将这一矩形所对应的相机图像的每一个像素进行透视逆变换，从而可以获得识别目标的原貌图像。
在实际应用中，相机的位置一旦发生变动，处理设备160会检测到已知几何多边形的顶点在相机图像中的位置发生了变动。系统100会根据新的顶点位置坐标根据以上所述方法重新计算透视变换矩阵，然后根据新的透视变换矩阵对相机图像进行透视校准，进而获得识别目标的原貌图像。这一过程也可以称为透视校准调整模式。由于这一过程可以是实时动态的，所以系统始终可以获得识别目标的原貌图像。这为后续的图像识别提供了可靠的保障。
图4示出了根据本公开的一个实施例的可视化矩形的示意图。如图4所示，可以在电气设备盘柜的边缘放置可视化的矩形框架，便于计算机对这一矩形框架进行识别和跟踪。例如，可视化特性可以通过特定颜色来指示。
图5示出了根据本公开的另一实施例的可视化矩形的示意图。如图5所示，可以在电气设备盘柜的边缘放置四个可视化的标签，使得这四个可视化的标签作为顶点形成矩形。例如，可视化特性可以通过特定颜色来指示。
通过将电气设备设置在可视化矩形内部，可以比较方便地界定电气设备的位置以便于对电气设备的状态进行识别。尽管图4和图5示出了在电气设备盘柜的边缘设置可视化矩形的两个实施例，但是预定义多边形也可以设置在电气设备盘柜或其他环境的其他位置处，只要满足电气设备在预定义多边形的平面内的条件即可。
总体而言，在获得校准后的现场设备原貌图像之后，可以使用计算机现场学习法对电气设备的状态进行快速识别。计算机现场学习法 是在系统启动图像识别前，预先建立目标识别配置数据库。例如，在校准后的现场设备原貌图像中，通过人机操作，逐个选取包含电气设备状态特征的图像区域，把这些区域的位置信息和所有反映电气设备状态变化特征图像都全部保存下来，形成目标识别配置数据库。
由于现场设备的安装位置固定不变，在系统进行识别时，系统读取目标识别配置数据库，根据数据库所指定的图像位置来搜索和匹配指定的电气设备特征图像，根据匹配结果来判定电气设备的数字量特征(如开、关、分、合等)。
如果电气设备具有模拟量特征(如指针仪表)，则在这局部范围内根据目标识别配置数据库所预定义的模拟量位置图像信息(例如，与模拟量的极大值和极小值对应的图像信息)、结合图像分析算法得出电气设备的模拟量数值。在识别过程中由于系统不需要对整个图像范围进行扫描判别，仅仅是对指定范围进行识别分析，所以可以实现快速判别。又由于每个设备状态的特征图像都单独定义，所以目标识别选择性强，误判率极低。
图6示出了根据本公开的一个实施例的识别电气设备的状态的方法600的流程图。图6可以由图1所示的系统100来实现，具体地，可以由系统100中的图像采集设备140和处理设备160来实现。
在602，图像采集设备140获取电气设备所处的环境的图像，其中电气设备设置在环境中的预定义多边形的平面上。该环境也可以称为电气设备现场或现场环境，如图1所示的电气设备现场120。在一些实施例中，预定义多边形是矩形，并且矩形的顶点的坐标已知。在一些实施例中，环境包括具有多边形的形状的可视化框架，例如，如图4所示。在一些实施例中，环境包括设置在所述多边形的顶点处的可视化标签，例如，如图5所示。
在604，处理设备160通过将所述图像中的所述预定义多边形还原成所述预定义多边形的原貌来获得所述电气设备在所述现场环境中的原貌图像。具体地，可以通过结合图2-图3所述的透视校准方法来获得原貌图像。
在606，处理设备160基于环境的原貌图像来确定电气设备的状态。这可以通过计算机现场学习法来实现。在一些实施例中，方法600还包括接收在原貌图像中对电气设备的选择，以确定电气设备在原貌图像中的位置，并将该位置作为电气设备的预定义位置。此外，还可以确定与电气设备处于第一状态相关联的第一特征图像。这一特征图像可以是原貌图像在电气设备的预定义位置处的一部分。
在电气设备的状态从第一状态改变到第二状态时，可以获取第二原貌图像，并确定与电气设备处于第二状态相关联的第二特征图像。状态的改变可以通过自动控制或人工操作来实现，本公开在此不受限制。例如，对于开关而言，第一状态可以是开状态，并且第二状态是关状态。在这种情况下，针对开状态和关状态分别获取相应的特征图像。
在一些实施例中，所述电气设备的状态具有模拟量特征，所述第一状态和所述第二状态对应于所述电气设备的模拟量特征的第一数值和第二数值，并且所述电气设备的状态的变化与所述第一数值和第二数值具有预定义关系。例如，对于指针类模拟量，第一数值和第二数值可以是模拟量的极大值和极小值。在这种情况下，模拟量特征的变化可以与两个数值具有预定义的线性或指数关系等。此外，在电气设备的温度测量中，可以使用不同的颜色来表示不同的温度。在这种情况下，第一数值和第二数值可以对应于两个不同的颜色或者温度。相应地，颜色或温度的变化可以与两个数值具有预定义的关系，例如可以由表格的形式来呈现。
以上介绍了方法300的现场学习模式的方法步骤。通常而言，学习模式是正常工作模式的基础。以下将结合模拟量和数字量分别介绍正常工作模式的若干实施例。
在一些实施例中，电气设备的状态具有数字量特征，确定电气设备的状态包括：获取电气设备的预定义位置和特征图像以及与特征图像对应的状态；在原貌图像内，基于预定义位置搜索与特征图像匹配的区域；以及基于与特征图像对应的状态来确定电气设备的状态。具 体而言，可以在预定义位置周围一定范围内来搜索与特征图像匹配的区域。以这种方式，可以不用对整个图像进行处理来确定电气设备的状态，而只需对与该电气设备最相关的图像部分进行处理，从而可以节省计算资源，提升计算效率。
在一些实施例中，电气设备的状态具有模拟量特征，并且确定电气设备的状态包括：获取所述电气设备的预定义位置和多个特征图像以及与所述多个特征图像分别对应的多个状态，所述多个状态对应于所述电气设备的模拟量特征的多个数值；在所述原貌图像内，基于预定义位置搜索与所述多个特征图像匹配的区域；在所述区域中识别所述模拟量特征的状态的指示；以及基于所述模拟量特征的状态的指示，通过所述电气设备的状态的变化与所述多个数值之间的预定义关系来确定所述电气设备的状态。如上所述，对于一些应用，多个状态可以对应于模拟量的极大值和极小值。为了清楚起见，在此不再赘述。
图7示出了根据本公开的一个实施例的电气设备状态识别装置700的框图。如图7所示，装置700包括相机模块760、主控模块740、图像识别模块710、目标识别配置数据库720、人机界面模块770、电气识别状态模块730、通信模块750和电源模块780。应当理解，以上功能模块的划分只是为了方便对图像识别方法和原理的说明。在具体应用中，多个模块可以合并在一个模块中，一个模块可以进一步拆分为多个模块，并且也可以省略其中的一个或多个模块。
相机模块760可以获取电气设备现场的图像，并且将图像数据传输给主控模块740。主控模块740将相机模块760获取的图像传输给图像识别模块710进行图像识别。图像识别模块710接收主控模块740传输来的图像数据，并结合目标识别配置数据库720的预定义信息进行图像校准和图像识别，并将数字化的识别结果传输给电气设备状态模块730。电气设备状态模块730保存现场电气设备的状态信息，并且在一些实施例中用数字化方式来存在电气设备的状态信息，以便于管理和传输。目标识别配置数据库740中的信息是用户使用人机界面模块770，通过人机操作对需要识别的目标图像进行计算机现场学习 而获得的。主控模块740读取电气设备状态模块730中的数据，并对数据进行处理。通信模块750将电气设备状态信息发送给远程服务器，以实现数字化、网络化的信息共享和决策分析。电源模块780可以为装置700中的模块提供电源。
以下结合图8进一步详细描述图7所示的装置700的原理，其中图8示出了根据本公开的一个实施例的识别电气设备的状态的方法800的流程图。
在802，主控模块740从相机模块760中获取电气设备现场的图像。电气设备现场包括一个或多个电气设备或部件，并且设置有具有预定义形状的多边形。
在804，主控模块740识别多边形的顶点在图像中的坐标，并对图像进行透视校准，获得现场设备的原貌图像。具体地，可以通过结合图2-图3所述的透视校准方法来获得原貌图像。
在806，主控模块740从人机界面模块770获取系统的当前工作模式，其中工作模式可以包括：透视校准调整模式、计算机现场学习模式和正常工作模式等。
如果在806确定为透视校准调整模式，则在810通过人机界面模块770获取多边形顶点的坐标定义，启动透视校准，并且在812将透视校准参数(例如，如上所述的透视变换矩阵)保存在目标识别配置数据库720中。然后，系统可以进入下一循环，返回702。
如果在806确定为计算机现场学习模式，则在820通过人机界面模块，在现场设备原貌图像中选定电气设备特征图像的位置和区域，定义每个设备状态与该特征图像的关系，建立目标识别配置数据库720。例如，如果电气设备是一个具有开和关这两种状态的开关，则针对开和关这两种状态分别获取现场的原貌图像，并分别选取与开关对应的特征图像的区域和与该特征图像相对应的状态，也称为目标识别参数。以这种方式，建立目标识别配置数据库720。然后，系统可以进入下一循环，返回702。
对于具有模拟量特征的电气设备，可以针对预定义范围内的极限 状态获取现场的原貌图像，并定义相应的模拟量数值。例如，对于仪表指针，可以针对仪表指针中的起始位置和终止位置来获取现场的原貌图像，并针对两个图像定义相应的模拟量数值。
如果在808确定为正常工作模式，则在830，图像识别模块710读取目标识别配置数据库720中的目标识别参数。在832，可以根据目标识别参数所指定的区域在原貌图像中搜索与指定目标相匹配的图像。如上所述，设备现场通常是保持固定不变的，因此，可以在指定区域进行搜索，而不用对整个图像进行搜索，从而具有更高的准确性并节省计算量。
如果在834确定图像匹配成功，则在836确定电气设备的状态信息。对于具有数字量特征的电气设备(例如，开关)，由于在目标识别数据库720中所匹配的图像具有相对应的状态，可以直接确定电气设备的状态。对于具有模拟量特征的电气设备，根据目标识别配置数据库720中定义的模拟量的极限状态的特征图像，结合图像分析算法得到设备的模拟量数值。例如，对于仪表指针，可以在匹配的区域中标识指示模拟量特征的指示符的位置，并且基于指示符的位置和数据库中定义的模拟量的极限状态的特征图像来确定电气设备的状态。此外，图像识别模块710可以根据识别结果更新电气设备状态模块730中的信息。
如果在834确定图像匹配识别不成功，则将错误信息保存在电气设备状态模块730中。主控模块740可以在840将电气设备状态模块730中的信息通过通信模块760发送给远程网络。然后，系统可以进入下一循环，返回702。
在这些实施例中，相机图像识别处理计算量小，识别效率高。此外，目标识别的性能稳定，具备更高的正确率和可靠性。方便用户对现场识别目标重新进行计算机现场学习、重新配置识别内容而不需要改变源程序。从而使本方法能适应各种现场设备的型号和布局，适应各种不同的照明环境。本方法具备自动适应相机位置和角度变化的功能，对相机位置变化的宽容性大。
以上结合图1-图8描述了根据本公开的一些实施例的用于识别电气设备的状态的方法、设备和系统。应当理解，本文中以上描述的功能可以至少部分地由一个或多个硬件逻辑部件来执行。例如，非限制性地，可以使用的示范类型的硬件逻辑部件包括：现场可编程门阵列(FPGA)、专用集成电路(ASIC)、专用标准产品(ASSP)、片上系统(SOC)、复杂可编程逻辑设备(CPLD)等等。
用于实施本公开的方法的程序代码可以采用一个或多个编程语言的任何组合来编写。这些程序代码可以提供给通用计算机、专用计算机或其他可编程数据处理装置的处理器或控制器，使得程序代码当由处理器或控制器执行时使流程图和/或框图中所规定的功能/操作被实施。程序代码可以完全在机器上执行、部分地在机器上执行，作为独立软件包部分地在机器上执行且部分地在远程机器上执行或完全在远程机器或服务器上执行。
在本公开的上下文中，机器可读介质可以是有形的介质，其可以包含或存储以供指令执行系统、装置或设备使用或与指令执行系统、装置或设备结合地使用的程序。机器可读介质可以是机器可读信号介质或机器可读储存介质。机器可读介质可以包括但不限于电子的、磁性的、光学的、电磁的、红外的、或半导体系统、装置或设备，或者上述内容的任何合适组合。机器可读存储介质的更具体示例会包括基于一个或多个线的电气连接、便携式计算机盘、硬盘、随机存取存储器(RAM)、只读存储器(ROM)、可擦除可编程只读存储器(EPROM或快闪存储器)、光纤、便捷式紧凑盘只读存储器(CD-ROM)、光学储存设备、磁储存设备、或上述内容的任何合适组合。
此外，虽然采用特定次序描绘了各操作，但是这应当理解为要求这样操作以所示出的特定次序或以顺序次序执行，或者要求所有图示的操作应被执行以取得期望的结果。在一定环境下，多任务和并行处理可能是有利的。同样地，虽然在上面论述中包含了若干具体实现细节，但是这些不应当被解释为对本公开的范围的限制。在单独的实现的上下文中描述的某些特征还可以组合地实现在单个实现中。相反地， 在单个实现的上下文中描述的各种特征也可以单独地或以任何合适的子组合的方式实现在多个实现中。
尽管已经采用特定于结构特征和/或方法逻辑动作的语言描述了本主题，但是应当理解所附权利要求书中所限定的主题未必局限于上面描述的特定特征或动作。相反，上面所描述的特定特征和动作仅仅是实现权利要求书的示例形式。

Claims (24)

  1. 一种用于识别电气设备的状态的方法，包括：
    获取所述电气设备在现场环境中的图像，其中所述电气设备设置在所述现场环境中的预定义多边形的平面上；
    通过将所述图像中的所述预定义多边形还原成所述预定义多边形的原貌来获得所述电气设备在所述现场环境中的原貌图像；以及
    基于所述电气设备在所述现场环境中的原貌图像，确定所述电气设备的状态。
  2. 根据权利要求1所述的方法，其中所述预定义多边形是矩形，并且所述矩形的顶点的坐标已知。
  3. 根据权利要求1所述的方法，其中所述现场环境包括以下至少一项：
    具有所述预定义多边形的形状的可视化框架；和
    设置在所述预定义多边形的顶点处的可视化标签。
  4. 根据权利要求1所述的方法，其中所述电气设备设置在预定义多边形内。
  5. 根据权利要求1所述的方法，还包括：
    接收在所述原貌图像中对所述电气设备的选择，以确定所述电气设备在所述原貌图像中的位置，并作为所述电气设备的预定义位置，并且确定与所述电气设备处于第一状态相关联的第一特征图像；以及
    响应于所述电气设备的状态从所述第一状态改变到所述第二状态，获取第二原貌图像并确定与所述电气设备处于第二状态相关联的第二特征图像。
  6. 根据权利要求5所述的方法，其中所述电气设备的状态具有模拟量特征，所述第一状态和所述第二状态对应于所述电气设备的模拟量特征的第一数值和第二数值，并且所述电气设备的状态的变化与所述第一数值和第二数值具有预定义关系。
  7. 根据权利要求1所述的方法，其中所述电气设备的状态具有数 字量特征，确定所述电气设备的状态包括：
    获取所述电气设备的预定义位置和特征图像以及与所述特征图像对应的状态；
    在所述原貌图像内，基于所述预定义位置搜索与所述特征图像匹配的区域；以及
    基于与所述特征图像对应的状态来确定所述电气设备的状态。
  8. 根据权利要求1所述的方法，其中所述电气设备的状态具有模拟量特征，并且确定所述电气设备的状态包括：
    获取所述电气设备的预定义位置和多个特征图像以及与所述多个特征图像分别对应的多个状态，所述多个状态对应于所述电气设备的模拟量特征的多个数值；
    在所述原貌图像内，基于所述预定义位置搜索与所述多个特征图像匹配的区域；
    在所述区域中识别所述模拟量特征的状态的指示；以及
    基于所述模拟量特征的状态的指示，通过所述电气设备的状态的变化与所述多个数值之间的预定义关系来确定所述电气设备的状态。
  9. 一种用于识别电气设备的状态的装置，包括：
    图像采集设备，被配置为获取所述电气设备在现场环境中的图像，其中所述电气设备设置在所述现场环境中的预定义多边形的平面上；
    处理设备，被配置为：
    通过将所述图像中的所述预定义多边形还原成所述预定义多边形的原貌来获得所述电气设备在所述现场环境中的原貌图像；以及
    基于所述电气设备在所述现场环境中的原貌图像，确定所述电气设备的状态。
  10. 根据权利要求9所述的装置，其中所述预定义多边形是矩形，并且所述矩形的顶点的坐标已知。
  11. 根据权利要求9所述的装置，其中所述现场环境包括以下至少一项：
    具有所述预定义多边形的形状的可视化框架；和
    设置在所述预定义多边形的顶点处的可视化标签。
  12. 根据权利要求9所述的装置，其中所述电气设备设置在预定义多边形内。
  13. 根据权利要求9所述的装置，其中所述处理设备还被配置为：
    接收在所述原貌图像中对所述电气设备的选择，以确定所述电气设备在所述原貌图像中的位置，并作为所述电气设备的预定义位置，并且确定与所述电气设备处于第一状态相关联的第一特征图像；以及
    响应于所述电气设备的状态从所述第一状态改变到所述第二状态，获取第二原貌图像并确定与所述电气设备处于第二状态相关联的第二特征图像。
  14. 根据权利要求13所述的装置，其中所述电气设备的状态具有模拟量特征，所述第一状态和所述第二状态指示所述电气设备的模拟量特征的第一数值和第二数值，所述电气设备的状态的变化与所述第一数值和第二数值具有预定义关系。
  15. 根据权利要求9所述的装置，其中所述电气设备的状态具有数字量特征，确定所述电气设备的状态包括：
    获取所述电气设备的预定义位置和特征图像以及与所述特征图像对应的状态；
    在所述原貌图像内，基于所述预定义位置搜索与所述特征图像匹配的区域；以及
    基于与所述特征图像对应的状态来确定所述电气设备的状态。
  16. 根据权利要求9所述的装置，其中所述电气设备的状态具有模拟量特征，并且确定所述电气设备的状态包括：
    获取所述电气设备的预定义位置和多个特征图像以及与所述多个特征图像分别对应的多个状态，所述多个状态对应于所述电气设备的模拟量特征的多个数值；
    在所述原貌图像内，基于所述预定义位置搜索与所述多个特征图像匹配的区域；
    在所述区域中识别所述模拟量特征的状态的指示；以及
    基于所述模拟量特征的状态的指示，通过所述电气设备的状态的变化与所述多个数值之间的预定义关系来确定所述电气设备的状态。
  17. 一种用于识别电气设备的状态的设备，包括：
    处理单元；以及
    存储器，耦合至所述处理单元并且包括存储于其上的指令，所述指令在由所述处理单元执行时使所述设备执行动作，所述动作包括：
    接收所述电气设备在现场环境中的图像，其中所述电气设备设置在所述现场环境中的预定义多边形的平面上；
    通过将所述图像中的所述预定义多边形还原成所述预定义多边形的原貌来获得所述电气设备在所述现场环境中的原貌图像；以及
    基于所述电气设备在所述现场环境中的原貌图像，确定所述电气设备的状态。
  18. 根据权利要求17所述的设备，其中所述动作还包括：
    接收在所述原貌图像中对所述电气设备的选择，以确定所述电气设备在所述原貌图像中的位置，并作为所述电气设备的预定义位置，并且确定与所述电气设备处于第一状态相关联的第一特征图像；
    响应于所述电气设备的状态从所述第一状态改变到所述第二状态，获取第二原貌图像并确定与所述电气设备处于第二状态相关联的第二特征图像。
  19. 根据权利要求17所述的设备，其中所述电气设备的状态具有数字量特征，并且其中确定所述电气设备的状态包括：
    获取所述电气设备的预定义位置和特征图像以及与所述特征图像对应的状态；
    在所述原貌图像内，基于所述预定义位置搜索与所述特征图像匹配的区域；以及
    基于与所述特征图像对应的状态来确定所述电气设备的状态。
  20. 根据权利要求17所述的设备，其中所述电气设备的状态具有 模拟量特征，并且确定所述电气设备的状态包括：
    获取所述电气设备的预定义位置和多个特征图像以及与所述多个特征图像分别对应的多个状态，所述多个状态对应于所述电气设备的模拟量特征的多个数值；
    在所述原貌图像内，基于所述预定义位置搜索与所述多个特征图像匹配的区域；
    在所述区域中识别所述模拟量特征的状态的指示；以及
    基于所述模拟量特征的状态的指示，通过所述电气设备的状态的变化与所述多个数值之间的预定义关系来确定所述电气设备的状态。
  21. 一种存储有指令的计算机可读存储介质，所述指令响应于由计算设备的执行而使得所述计算设备执行动作，所述动作包括：
    接收电气设备在现场环境中的图像，其中所述电气设备设置在所述现场环境中的预定义多边形的平面上；
    将所述图像中的所述预定义多边形还原成所述预定义多边形的原貌来获得所述电气设备在所述现场环境中的原貌图像；以及
    基于所述电气设备在现场环境中的原貌图像，确定所述电气设备的状态。
  22. 根据权利要求21所述的计算机可读存储介质，其中所述动作还包括：
    接收在所述原貌图像中对所述电气设备的选择，以确定所述电气设备在所述原貌图像中的位置，并作为所述电气设备的预定义位置，并且确定与所述电气设备处于第一状态相关联的第一特征图像；
    响应于所述电气设备的状态从所述第一状态改变到所述第二状态，获取第二原貌图像并确定与所述电气设备处于第二状态相关联的第二特征图像。
  23. 根据权利要求21所述的计算机可读存储介质，其中所述电气设备的状态具有数字量特征，并且其中确定所述电气设备的状态包括：
    获取所述电气设备的预定义位置和特征图像以及与所述特征图像对应的状态；
    在所述原貌图像内，基于所述预定义位置搜索与所述特征图像匹配的区域；以及
    基于与所述特征图像对应的状态来确定所述电气设备的状态。
  24. 根据权利要求21所述的计算机可读存储介质，其中所述电气设备的状态具有模拟量特征，并且确定所述电气设备的状态包括：
    获取所述电气设备的预定义位置和多个特征图像以及与所述多个特征图像分别对应的多个状态，所述多个状态对应于所述电气设备的模拟量特征的多个数值；
    在所述原貌图像内，基于所述预定义位置搜索与所述多个特征图像匹配的区域；
    在所述区域中识别所述模拟量特征的状态的指示；以及
    基于所述模拟量特征的状态的指示，通过所述电气设备的状态的变化与所述多个数值之间的预定义关系来确定所述电气设备的状态。
PCT/CN2018/108037 2017-09-30 2018-09-27 用于识别电气设备的状态的方法和装置 Ceased WO2019062818A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP18860822.8A EP3678048A4 (en) 2017-09-30 2018-09-27 METHOD AND DEVICE FOR IDENTIFYING THE CONDITION OF AN ELECTRICAL DEVICE
US16/652,102 US11302099B2 (en) 2017-09-30 2018-09-27 Method and device for recognizing states of electrical devices

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710917722.7 2017-09-30
CN201710917722.7A CN109598177B (zh) 2017-09-30 2017-09-30 用于识别电气设备的状态的方法和装置

Publications (1)

Publication Number Publication Date
WO2019062818A1 true WO2019062818A1 (zh) 2019-04-04

Family

ID=65900846

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/108037 Ceased WO2019062818A1 (zh) 2017-09-30 2018-09-27 用于识别电气设备的状态的方法和装置

Country Status (4)

Country Link
US (1) US11302099B2 (zh)
EP (1) EP3678048A4 (zh)
CN (1) CN109598177B (zh)
WO (1) WO2019062818A1 (zh)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112085654A (zh) * 2020-08-10 2020-12-15 国网上海市电力公司 一种可配置式模拟屏识别系统

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP4033397B1 (en) * 2021-01-21 2025-11-26 ABB S.p.A. A computer-implemented method for assisting a user in interacting with an electronic relay for electric power distribution grids
US20230011330A1 (en) * 2021-07-09 2023-01-12 At&T Intellectual Property I, L.P. Device condition determination
CN114359844B (zh) * 2022-03-21 2022-06-21 广州银狐科技股份有限公司 一种基于色彩识别的aed设备状态监测方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103324943A (zh) * 2013-06-18 2013-09-25 中国人民解放军第二炮兵工程大学 一种复杂设备面板图像多子区状态识别方法
CN105825204A (zh) * 2016-04-08 2016-08-03 国家电网公司 一种智能相机夜间识别电力设备表计的方法
CN105894015A (zh) * 2016-03-28 2016-08-24 国网浙江省电力公司宁波供电公司 一种道闸状态分析方法和系统
CN106355187A (zh) * 2016-09-07 2017-01-25 西华大学 视觉信息在电力设备监测中的应用

Family Cites Families (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2006011141A2 (en) * 2004-07-25 2006-02-02 Israel Aerospace Industries Ltd. Method and system for the acquisition of data and for the display of data
FR2990286B1 (fr) * 2012-05-07 2019-09-06 Schneider Electric Industries Sas Procede d'affichage en realite augmentee d'une information relative a un equipement cible sur un ecran d'un dispositif electronique, produit programme d'ordinateur, dispositif electronique et equipement electrique associes
US9122921B2 (en) * 2013-06-12 2015-09-01 Kodak Alaris Inc. Method for detecting a document boundary
US9123111B2 (en) 2013-08-15 2015-09-01 Xerox Corporation Methods and systems for detecting patch panel ports from an image in which some ports are obscured
ITBO20130466A1 (it) * 2013-08-29 2015-03-01 Umpi Elettronica Societa A Respo Nsabilita Lim Metodo di ispezione e/o manutenzione di una parte di un impianto industriale mediante realta' aumentata, e corrispondente sistema per guidare l'ispezione e/o la manutenzione della parte di impianto industriale
US9576348B2 (en) 2014-11-14 2017-02-21 Adobe Systems Incorporated Facilitating text identification and editing in images
US9940521B2 (en) * 2015-02-27 2018-04-10 Sony Corporation Visibility enhancement devices, systems, and methods
JP6670585B2 (ja) * 2015-10-30 2020-03-25 Juki株式会社 管理装置
CN105930837A (zh) * 2016-05-17 2016-09-07 杭州申昊科技股份有限公司 基于自主巡检机器人的变电站仪表设备图像识别方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103324943A (zh) * 2013-06-18 2013-09-25 中国人民解放军第二炮兵工程大学 一种复杂设备面板图像多子区状态识别方法
CN105894015A (zh) * 2016-03-28 2016-08-24 国网浙江省电力公司宁波供电公司 一种道闸状态分析方法和系统
CN105825204A (zh) * 2016-04-08 2016-08-03 国家电网公司 一种智能相机夜间识别电力设备表计的方法
CN106355187A (zh) * 2016-09-07 2017-01-25 西华大学 视觉信息在电力设备监测中的应用

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3678048A4 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112085654A (zh) * 2020-08-10 2020-12-15 国网上海市电力公司 一种可配置式模拟屏识别系统

Also Published As

Publication number Publication date
US11302099B2 (en) 2022-04-12
CN109598177B (zh) 2023-12-01
CN109598177A (zh) 2019-04-09
US20200250431A1 (en) 2020-08-06
EP3678048A4 (en) 2021-05-05
EP3678048A1 (en) 2020-07-08

Similar Documents

Publication Publication Date Title
US11145080B2 (en) Method and apparatus for three-dimensional object pose estimation, device and storage medium
CA2774974C (en) Real time hand tracking, pose classification, and interface control
US9639758B2 (en) Method and apparatus for processing image
WO2019062818A1 (zh) 用于识别电气设备的状态的方法和装置
KR101374139B1 (ko) 감시시스템의 영상 융합을 통한 객체 감시 방법
US11989928B2 (en) Image processing system
CN111523423B (zh) 一种电力设备识别方法及装置
GB2512460A (en) Position and orientation measuring apparatus, information processing apparatus and information processing method
CN109919128B (zh) 控制指令的获取方法、装置及电子设备
US20210374977A1 (en) Method for indoor localization and electronic device
US20170109932A1 (en) Content projection apparatus, content projection method, and computer readable storage medium
US20160005156A1 (en) Infrared selecting device and method
JP2025078813A (ja) 情報処理装置、情報処理方法、及びプログラム
KR20220015056A (ko) 3d 환경 모델을 이용하여 피사체의 실제 크기를 측정 가능한 cctv 관제 시스템
CN112153269B (zh) 应用于电子设备的图片显示方法、装置、介质与电子设备
CN110909685A (zh) 姿势估计方法、装置、设备及存储介质
KR20210114838A (ko) 멀티퍼슨 체온 측정 방법, 장치, 전자 기기 및 저장 매체
JP2014106597A (ja) 自律移動体、物体情報取得装置、および物体情報取得方法
CN116363583B (zh) 一种用于顶视角的人体识别方法、装置、设备及介质
CN112097742A (zh) 一种位姿确定方法及装置
KR101517538B1 (ko) 중심 가중치 맵을 이용한 중요도 영역 검출 장치 및 방법, 이를 위한 프로그램을 기록한 기록 매체
JP7074174B2 (ja) 識別器学習装置、識別器学習方法およびコンピュータプログラム
JPWO2017126072A1 (ja) 画像認識システム、カメラ状態推定装置及び記憶媒体
CN119967122A (zh) 巡检控制方法、装置及电子设备
CN103900713A (zh) 热像检测装置和热像检测方法

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: 18860822

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2018860822

Country of ref document: EP

Effective date: 20200330