A method and system for identifying platen status and row and column numbers

Abstract

The present invention belongs to the field of image recognition technology, and provides a method and system for identifying the state and row and column number of a pressure plate. First, a pressure plate image is obtained, and the pressure plate image is filtered using an image quality filter and an image reflection filter. If the pressure plate image passes the filtering, it is used as a standard image, otherwise, the recognition process is directly terminated; then, the pressure plate gathering area in the standard image is identified; the state and position of each pressure plate in the pressure plate gathering area are identified; finally, based on the position of each pressure plate, the row and column number of each pressure plate is identified. The effective identification of the state of the pressure plate switch in the protection screen cabinet and its row and column number is achieved, the recognition accuracy is improved, the intelligent application of power business is promoted, and the burden on grassroots power employees is effectively reduced.

Description

Technical Field

[0001] The invention belongs to the technical field of image recognition, and in particular relates to a method and system for recognizing a pressing plate state and row and column numbers. Background Art

[0002] The statements in this section merely provide background information related to the present invention and do not necessarily constitute prior art.

[0003] There are usually 10-20 protection cabinets in a 110kv substation, and the number of protection cabinets in 220kv and 500kv substations can reach hundreds. Each protection cabinet has as many as fifty-four pressure plate switches, and inspection work requires inspection of these pressure plate switches.

[0004] However, the current manual inspection process has problems such as heavy workload, poor timeliness, and paper-based work records. In addition, due to factors such as physical and psychological quality, sense of responsibility, technical level, and work experience, inspectors may make mistakes, leaving safety hazards and even causing major safety disasters. How to locate the pressure plate, how to identify the pressure plate status, and how to check the correctness of the status are the key points in the inspection process.

[0005] There are many methods to solve the problem of identifying the status of protective pressure plates, but none of them are carried out in conjunction with the entire inspection process. They only identify the status of the pressure plates, but do not know the location of the pressure plates, and do not perform automated verification. Only by solving the core problems in the entire process can we improve intelligence and effectively reduce the burden on power workers. Summary of the invention

[0006] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a method and system for identifying the status of a pressure plate and the row and column numbers, which can automatically identify the position of each pressure plate and the row and column numbers of the pressure plate, effectively reducing the burden on power workers.

[0007] In order to achieve the above object, the present invention adopts the following technical solution:

[0008] A first aspect of the present invention provides a method for identifying a platen state and row and column numbers, comprising:

[0009] Acquire a platen image;

[0010] The image quality filter and the image reflection filter are used to filter the platen image. If the platen image passes the filtering, it is used as the standard image. Otherwise, the recognition process is terminated directly.

[0011] Identify the platen aggregation area in the standard image;

[0012] Identify the status and location of each platen in the platen aggregation area;

[0013] Based on the position of each platen, each platen row number is identified.

[0014] Furthermore, the specific steps of identifying the row and column number of each pressing plate are:

[0015] Based on the position of each platen, the row and column deviation thresholds and the center point of each platen are calculated;

[0016] Based on the row deviation threshold and the center point of each platen, the platens are aggregated to obtain multiple groups, and the row means of each group are sorted to obtain the row number of each group. The row numbers of all platens in the group are the row numbers of the corresponding group.

[0017] Based on the center point of each platen, the platens in each group are sorted to obtain the column number of each platen;

[0018] Based on the column deviation threshold, the column number of each platen is corrected.

[0019] Furthermore, the specific calculation method of the row and column deviation threshold is:

[0020] The width of each platen is counted, the average width of the platen is calculated, and then divided by 2 to obtain the column deviation threshold;

[0021] The height of each platen is counted, the average height of the platen is calculated, and then divided by 2 to obtain the row deviation threshold.

[0022] Furthermore, the specific steps of polymerizing the pressing plate are:

[0023] (a) taking out a pressing plate from the ungrouped pressing plates;

[0024] (b) calculating the aggregate range of each ungrouped platen based on the center point of each ungrouped platen and the row deviation threshold;

[0025] (c) using the center point of the removed pressing plate to compare with the aggregation range of each ungrouped pressing plate in turn, if the center point of the removed pressing plate is within the aggregation range of a certain ungrouped pressing plate, then the ungrouped pressing plate and the removed pressing plate are grouped together;

[0026] (d) Return to step (a) until there are no ungrouped press plates left.

[0027] Furthermore, the specific steps of correcting the column number of each pressing plate are:

[0028] Calculate the number of pressure plates in each group and find the group with the most pressure plates as the reference group;

[0029] The center point position of each to-be-corrected pressing plate of the non-reference group and the column deviation threshold constitute the detection range;

[0030] When the center point of a certain reference group pressure plate falls within a certain detection range, the column number of the pressure plate to be calibrated corresponding to the detection range is the column number of the reference group pressure plate.

[0031] Furthermore, the specific steps of filtering by the image quality filter are:

[0032] Calculate the difference between the RGB mean of the input image and the mean of the image grayscale;

[0033] When the difference exceeds the set mean difference, it means that the image quality is poor and the recognition process ends directly; otherwise, the image is a high-quality image.

[0034] Furthermore, the specific steps of filtering by the image reflection filter are:

[0035] Calculate the mean and variance of the grayscale image of the high-quality image minus the brightness mean;

[0036] The deviation value is calculated based on the mean and variance of the grayscale image of the high-quality image minus the brightness mean;

[0037] When the deviation value is greater than the set threshold, it indicates that the brightness is abnormal and the recognition process ends directly; otherwise, the image is a standard image.

[0038] A second aspect of the present invention provides a platen state and row and column number recognition system, comprising:

[0039] An image acquisition module is configured to: acquire a pressure plate image;

[0040] A filtering module is configured to: filter the platen image using an image quality filter and an image reflection filter, and if the platen image passes the filtering, use it as a standard image; otherwise, directly end the recognition process;

[0041] A platen gathering area recognition module is configured to: recognize a platen gathering area in a standard image;

[0042] a platen status and position identification module configured to: identify the status and position of each platen in the platen aggregation area;

[0043] The platen row and column number identification module is configured to: identify each platen row and column number based on the position of each platen.

[0044] A third aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps in the method for identifying a platen state and row and column numbers as described above.

[0045] The fourth aspect of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps in the method for identifying the pressure plate status and row and column numbers as described above are implemented.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] The present invention provides a method for identifying the state of a pressure plate and its row and column numbers. The method performs image recognition on the area where the pressure plate is located based on deep learning neural network algorithm modeling. The method can effectively identify the state of the pressure plate switch in the protection screen cabinet and its row and column number, thereby improving the recognition accuracy.

[0048] The present invention provides a method for identifying a platen state and row and column numbers, which ensures the robustness of identification and good adaptability to different environments and scenarios through a variety of technologies and processes.

[0049] The present invention provides a method for identifying the status of a pressure plate and row and column numbers. The identification of row and column numbers facilitates integration with business modules, accelerates implementation, improves the intelligent application of power business, and effectively reduces the burden on grassroots power employees. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The accompanying drawings in the specification, which constitute a part of the present invention, are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute improper limitations on the present invention.

[0051] Figure 1 This is an overall flow chart of a method for identifying a platen state and row and column numbers according to a first embodiment of the present invention;

[0052] Figure 2 This is a flow chart of the pressure plate position identification in the first embodiment of the present invention. DETAILED DESCRIPTION

[0053] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0054] It should be noted that the following detailed descriptions are all illustrative and intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meanings as those commonly understood by those skilled in the art to which the present invention belongs.

[0055] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprising" and / or "including" are used in this specification, it indicates the presence of features, steps, operations, devices, components and / or combinations thereof.

[0056] Embodiment 1

[0057] like Figure 1-2 As shown, this embodiment provides a method for identifying the state of a platen and row and column numbers, which mainly consists of two parts: platen state identification and platen row and column number identification. Platen state identification includes: image quality filter, image reflection filter, platen area identification, platen state identification and position detection; platen row and column number identification includes: calculating the center point of each platen, counting row and column deviation thresholds, classifying and sorting objects of the same row, sorting each group of row objects by column, and identifying the row and column numbers of platen objects; the specific steps are as follows:

[0058] Step 1: Obtain a pressure plate image, wherein the pressure plate image is a screen cabinet pressure plate image.

[0059] Step 2: Use image quality filter and image reflection filter to filter the platen image. If the platen image passes the filtering, it is used as the standard image. Otherwise, the recognition process ends directly.

[0060] Step 201: Use an image quality filter to filter the quality of the platen image. If the platen image passes the filter, it is regarded as a high-quality image. Otherwise, the recognition process ends directly.

[0061] Among them, the image quality filter first calculates the RGB mean (meanValue) of the input pressure plate image and the mean of the image grayscale (grayMeanValue); then calculates the difference between the RGB mean of the input pressure plate image and the mean of the image grayscale. When the difference exceeds the set mean difference, that is, meanValue-grayMeanValue>threshold, it means that the image quality is poor, and the recognition process is directly terminated and the reason is fed back. Among them, threshold is the set hyperparameter, that is, the mean difference, and the default value is 3; otherwise, the image quality is good, the input image is a high-quality image, and the following process continues.

[0062] Step 202: Use an image reflection filter to filter the reflection phenomenon of the high-quality image. If the high-quality image passes the filter, it is used as the standard image. Otherwise, the recognition process ends directly.

[0063] Among them, the image reflection filter first sets the hyperparameter (brightness mean), the default value is 128, and the brightness mean is subtracted from the value of each pixel on the grayscale image of the input high-quality image to obtain the processed image; the mean and variance of the processed image are calculated. When the image has reflection, the brightness will be abnormal, the mean will deviate from the mean point, and the variance will be small; when the calculated deviation value is greater than the set threshold 1, that is, the absolute value of the ratio of the mean to the variance, it means that the brightness is abnormal, and the recognition process is directly terminated and the reason is fed back; otherwise, the brightness is normal, the input image is a standard image, and the following process will continue.

[0064] Step 3: Identify the platen clustering area in the standard image. Specifically, the platen clustering area recognition model is used to identify the position of the platen clustering area, which is conducive to narrowing the target range and eliminating interference. After recognition, the image outside the area is cropped and used as the input for the next step.

[0065] The training of the platen cluster area recognition model requires the preparation of a sample set, and the platen cluster area is annotated with a rectangular box (x, y, w, h), where x and y are the x value and y value of the point closest to the origin of the platen cluster area, w is the width of the platen cluster area, and h is the height of the platen cluster area. A category is marked (platen_area); the platen switch cluster area in the collected sample image is annotated in Pascal VOC format using an annotation tool, and each image corresponds to an annotation file (XML).

[0066] As an implementation method, a training set, a validation set, and a test set are set in a ratio of 7:2:1, and a deep learning neural network algorithm is used to establish a pressure plate clustering area recognition model.

[0067] Step 4: Identify the state and position of each platen in the platen clustering area. Specifically, a platen state and position recognition model is used to recognize the state and position of each platen in the image recognized through the platen area.

[0068] The training of the platen status and position recognition model requires sample data. Use a rectangular box to annotate each platen (x, y, w, h), where x and y are the x and y values ​​of the point closest to the origin in the platen area, w is the width of the platen area, and h is the height of the platen area. The platen status is also marked: (open|close). Use the annotation tool to annotate the status of each platen in the collected sample image in Pascal VOC format, and each image corresponds to an annotation file (XML).

[0069] As an implementation method, a training set, a validation set, and a test set are set in a ratio of 7:2:1, and a deep learning neural network algorithm is used to establish a platen state and position recognition model.

[0070] Step 5: Based on the position of each platen, identify the row and column number of each platen, such as Figure 2 As shown, the specific steps of platen row and column number identification include:

[0071] Step 501: Calculate the center point of each platen based on the position of each platen: Calculate the center point of each platen through the position coordinate information (x, y, w, h) of each platen, and obtain the center point information (x c ,y c ), the calculation formula is x c =(x+w) / 2,y c =(y+h) / 2.

[0072] Step 502, based on the position of each pressure plate, count the row and column deviation thresholds: count the width w of each pressure plate, calculate the average width of the pressure plate, and then divide it by 2 to get the column deviation threshold; count the height h of each pressure plate, calculate the average height of the pressure plate, and then divide it by 2 to get the row deviation threshold.

[0073] Step 503, classify and sort the objects of the same row. Specifically, based on the row deviation threshold and the center point of each pressure plate, aggregate the pressure plates to obtain multiple groups, and sort the row means of each group to obtain the row number of each group. The row numbers of all pressure plates in the group are the row numbers of the corresponding groups.

[0074] (1) Use the y value of the center point of the platen c The deviation threshold of values ​​and rows is aggregated. The aggregation method is:

[0075] (a) taking out a pressing plate from the ungrouped pressing plates;

[0076] (b) Based on the center point of each ungrouped platen and the row deviation threshold, the aggregation range of each ungrouped platen is calculated. The aggregation range is the center point y of each ungrouped platen. c Deviation threshold for value ± row;

[0077] (c) Use the center point y of the removed pressure plate c The value is compared with the aggregation range of each ungrouped pressure plate in turn. If the center point y of the taken pressure plate c If the value is within the aggregation range of a certain ungrouped platen, the ungrouped platen is grouped with the removed platen and compared with all ungrouped platens;

[0078] (d) Again, take out one of the ungrouped platens, return to step (a), and execute the above steps until there are no ungrouped platens left.

[0079] (2) After calculating the aggregation, the center point y of all the pressure plates in each group cThe values ​​are accumulated to get the row mean, and the row number of each group is obtained by sorting the row means in ascending order. All the pressure plates in the group have corresponding row numbers.

[0080] Step 504: sort the pressure plates in each group based on the center point of each pressure plate to obtain the column number of each pressure plate. Specifically, each group of row objects is sorted by column, and the pressure plates in each group are sorted by the x value of the center point. c Sort the values ​​in ascending order to obtain the column number of each pressure plate.

[0081] Step 505: Based on the column deviation threshold, the column number of each platen is corrected. Specifically, the row and column numbers of the platen objects are identified, the number of platens in each group is calculated, and the group containing the most platens is found as the reference group. The column number of the platen objects in the reference group is used as the reference target; the center point position of each platen to be corrected in other groups (non-reference groups) and the column deviation threshold constitute the detection range (the x value of the center point of the platen to be corrected). c value ± the deviation threshold of the column), using the x value of the center point of the platen to be corrected. c The deviation threshold of the value ± column from the center point x of each platen of the reference group c When the center point x of a reference group platen is c The value falls within the detection range of a certain pressure plate to be corrected (the x value of the center point of the pressure plate to be corrected). c If the column number of the platen to be corrected is within the detection range ± the deviation threshold of the column, then the column number of the platen to be corrected corresponding to the detection range is the column number of the platen in the reference group, and the correct column number is obtained. The main purpose is to correct the column number, because some platen objects may not be identified. In the same group, if the column number of a platen is wrong, the column numbers of the following platens are also wrong.

[0082] The method for identifying the pressure plate status and row and column numbers provided in this embodiment only requires the user to input an image of a screen cabinet, and can automatically identify the position (row and column number) and status (on or off) of each hard pressure plate.

[0083] Embodiment 2

[0084] This embodiment provides a platen status and row and column number recognition system, which specifically includes the following modules:

[0085] An image acquisition module is configured to: acquire a pressure plate image;

[0086] A filtering module is configured to: filter the platen image using an image quality filter and an image reflection filter, and if the platen image passes the filtering, use it as a standard image; otherwise, directly end the recognition process;

[0087] A platen gathering area recognition module is configured to: recognize a platen gathering area in a standard image;

[0088] a platen status and position identification module configured to: identify the status and position of each platen in the platen aggregation area;

[0089] The platen row and column number identification module is configured to: identify each platen row and column number based on the position of each platen.

[0090] It should be noted here that each module in this embodiment corresponds to each step in Example 1 one by one, and the specific implementation process is the same, which will not be repeated here.

[0091] Embodiment 3

[0092] This embodiment provides a computer-readable storage medium on which a computer program is stored. When the program is executed by a processor, the steps in the method for identifying a platen state and row and column numbers as described in the first embodiment above are implemented.

[0093] Embodiment 4

[0094] This embodiment provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the program, the steps in the method for identifying a pressure plate state and row and column numbers as described in the first embodiment above are implemented.

[0095] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of hardware embodiments, software embodiments, or embodiments combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0096] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0097] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0098] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0099] A person skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc.

[0100] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for identifying a platen state and row and column numbers, characterized in that: include: Acquire a platen image; The image quality filter and the image reflection filter are used to filter the platen image. If the platen image passes the filtering, it is used as the standard image. Otherwise, the recognition process is terminated directly. Identify the platen aggregation area in the standard image; Identify the status and location of each platen in the platen aggregation area; Based on the position of each platen, identify each platen row and column number; The specific steps of filtering by the image quality filter are: Calculate the difference between the RGB mean of the input image and the mean of the image grayscale; When the difference exceeds the set mean difference, it means that the input image quality is poor and the recognition process ends directly; otherwise, the input image is a high-quality image; The specific steps of filtering by the image reflection filter are: Calculate the mean and variance of the grayscale image of the high-quality image minus the brightness mean; The deviation value is calculated based on the mean and variance of the grayscale image of the high-quality image minus the brightness mean; When the deviation value is greater than the set threshold, it indicates that the brightness is abnormal and the recognition process ends directly; otherwise, the image is a standard image.

2. A method for identifying a platen state and row and column numbers as claimed in claim 1, characterized in that: The specific steps of identifying the row and column number of each pressing plate are: Based on the position of each platen, the row and column deviation threshold and the center point of each platen are calculated; the specific calculation method of the row and column deviation threshold is: The width of each platen is counted, the average width of the platen is calculated, and then divided by 2 to obtain the column deviation threshold; The height of each platen is counted, the average height of the platen is calculated, and then divided by 2 to obtain the row deviation threshold; Based on the row deviation threshold and the center point of each platen, the platens are aggregated to obtain multiple groups, and the row means of each group are sorted to obtain the row number of each group. The row numbers of all platens in the group are the row numbers of the corresponding group. Based on the center point of each platen, the platens in each group are sorted to obtain the column number of each platen; Based on the column deviation threshold, the column number of each platen is corrected.

3. A method for identifying a platen state and row and column numbers as claimed in claim 2, characterized in that: The specific steps of polymerizing the pressing plate are: (a) taking out a pressing plate from the ungrouped pressing plates; (b) calculating the aggregate range of each ungrouped platen based on the center point of each ungrouped platen and the row deviation threshold; (c) using the center point of the removed pressing plate to compare with the aggregation range of each ungrouped pressing plate in turn, if the center point of the removed pressing plate is within the aggregation range of a certain ungrouped pressing plate, then the ungrouped pressing plate and the removed pressing plate are grouped together; (d) Return to step (a) until there are no ungrouped press plates left.

4. A method for identifying a platen state and row and column numbers as claimed in claim 2, characterized in that: The specific steps of correcting the column number of each pressing plate are: Calculate the number of pressure plates in each group and find the group with the most pressure plates as the reference group; The center point position of each to-be-corrected pressing plate of the non-reference group and the column deviation threshold constitute the detection range; When the center point of a certain reference group pressure plate falls within a certain detection range, the column number of the pressure plate to be calibrated corresponding to the detection range is the column number of the reference group pressure plate.

5. A system for identifying the state of a platen and the number of rows and columns, characterized in that: include: An image acquisition module is configured to: acquire a pressure plate image; The filtering module is configured to: filter the platen image using an image quality filter and an image reflection filter. If the platen image passes the filtering, it is used as a standard image; otherwise, the recognition process is directly terminated. The specific steps of filtering by the image quality filter are: Calculate the difference between the RGB mean of the input image and the mean of the image grayscale; When the difference exceeds the set mean difference, it means that the input image quality is poor and the recognition process ends directly; otherwise, the input image is a high-quality image; The specific steps of filtering by the image reflection filter are: Calculate the mean and variance of the grayscale image of the high-quality image minus the brightness mean; The deviation value is calculated based on the mean and variance of the grayscale image of the high-quality image minus the brightness mean; When the deviation value is greater than the set threshold, it indicates that the brightness is abnormal and the recognition process ends directly; otherwise, the image is a standard image; A platen gathering area recognition module is configured to: recognize a platen gathering area in a standard image; a platen status and position identification module configured to: identify the status and position of each platen in the platen aggregation area; The platen row and column number identification module is configured to: identify each platen row and column number based on the position of each platen.

6. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps in a method for identifying a platen state and row and column numbers as described in any one of claims 1 to 4 are implemented.

7. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the steps in the method for identifying the platen status and row and column numbers as described in any one of claims 1-4 are implemented.

Images