A cabinet indicator light recognition method, device and storage medium based on template matching

Abstract

The present application discloses a cabinet indicator light recognition method based on template matching, which relates to the technical field of instrument light recognition, and solves the problems of many wrong detections and missed detections, high manual inspection costs, and the like in the prior art; the method comprises: obtaining a template image, obtaining a monitoring area, and determining a shape template matching image; obtaining a template position image, and sorting the template position image in order to determine a position indication image; the template position image is processed by an image processing tool on the shape template matching image to determine a digital image including indicator lights and digital tubes; obtaining a picture to be tested, and determining a picture of a region to be tested; comparing the picture of the region to be tested with the digital image to determine whether they match; if the matching result is yes, color processing is performed on the indicator lights in the picture to be tested, and a result array and a confidence array of the picture of the indicator lights to be tested are determined; according to the result array, a result image is determined; intelligent identification of indicator lights is realized, and the problem of wrong detection and missed detection will not occur, and a large amount of labor costs are saved.

Description

Technical Field

[0001] The present application relates to the technical field of instrument light recognition, and in particular to a cabinet indicator light recognition method, device and storage medium based on template matching. Background Art

[0002] To check whether the machines in the cabinet can work normally, it is necessary to inspect the indicator lights in the cabinet to determine whether the machines are working normally and take pre-processing measures for machines that are not working properly.

[0003] At present, the inspection of indicator lights on cabinets is generally carried out manually, that is, staff enter the interior to identify and register the indicator lights.

[0004] When faced with a high-risk, harsh working environment, it is difficult for workers to conduct inspections, and the inspection frequency is not high enough, which leads to the failure to find possible faults in time, affecting the judgment of whether the equipment is working properly. When manually inspecting the cabinet indicator lights, it is easy to make mistakes and miss inspections, and the cost of manual inspection is high, which also has a certain impact on the personal safety and health of workers. Summary of the invention

[0005] The embodiment of the present application solves the problems of false detection and missed detection during manual inspection of cabinets in the prior art, as well as inadequate inspection and high cost of inspection of cabinets in high-risk and harsh environments, by providing a cabinet indicator light recognition method based on template matching.

[0006] In a first aspect, an embodiment of the present invention provides a cabinet indicator light recognition method based on template matching, the method comprising:

[0007] Select the indicator light image of the cabinet to be tested, obtain the area to be monitored, and determine the shape template matching file;

[0008] Get the cabinet template diagram, and use image processing tools to determine the template location diagram including indicator lights and digital tubes;

[0009] Sorting the template position diagrams in order to determine a position indication diagram;

[0010] Manually marking the position indication map to determine a marking color position indication map;

[0011] Determine the characteristic data of the hue, saturation and brightness of the indicator light in the color position indication diagram using a color model;

[0012] The characteristic data is saved in a file, and a characteristic data file is determined;

[0013] The shape template matching file, the color position indication diagram and the image to be tested are compared and operated to determine a result array of the image to be tested;

[0014] Determine a confidence array according to the values ​​of the result array;

[0015] A result map is determined based on the confidence array.

[0016] In combination with the first aspect, in a possible implementation, determining the shape template matching map includes: selecting four or more feature graphics on the template image to determine the area to be monitored, and the feature graphics are included in the template matching map.

[0017] In combination with the first aspect, in a possible implementation, sorting the template position diagram in order includes: using a scanning method from left to right and from top to bottom to sequentially number the indicator lights and digital tubes in the template position diagram.

[0018] In combination with the first aspect, in a possible implementation manner, the color processing of the indicator light of the image to be tested includes: using the HSV color model to define a range of required colors.

[0019] In combination with the first aspect, in a possible implementation, the color processing of the indicator light of the image to be tested further includes: using color distance debugging;

[0020] The color distance debugging is to mark the comparison distance value of each indicator light on the indicator light of the image to be tested, and return the calculation result according to the comparison distance value.

[0021] In combination with the first aspect, in a possible implementation manner, the confidence array includes:

[0022] When color processing is performed on the image to be tested, a confidence value is determined according to the color distance.

[0023] In a second aspect, an embodiment of the present invention provides a cabinet indicator light recognition device based on template matching, the device comprising:

[0024] Shape template determination unit: used to obtain the template image, obtain the monitoring area, and determine the shape template matching map;

[0025] The template position determination unit is used to obtain a template position map, and sort the template position map in order to determine a position indication map; the template position map is processed by an image processing tool on the shape template matching map to determine a digital map including indicator lights and digital tubes;

[0026] Input unit: used to obtain the image to be tested and determine the image of the area to be tested;

[0027] Area matching unit: used to compare the image of the area to be tested with the digital image to determine whether they match;

[0028] A processing result generating unit: used for judging that if the matching result is yes, then color processing is performed on the indicator light of the image to be tested, and determining the result array and confidence array of the image of the indicator light to be tested;

[0029] A result graph generating unit is used to determine a result graph according to the result array.

[0030] In conclusion of the second aspect, in a possible implementation, the template position determination unit further includes: selecting four or more feature graphics on the template image to determine the area to be monitored, and the feature graphics are included in the template matching image.

[0031] In conclusion of the second aspect, in a possible implementation, the template position sorting unit also includes: sorting the template position diagram in order, including: using a scanning method from left to right and from top to bottom to sequentially number the indicator lights and digital tubes in the template position diagram.

[0032] In combination with the second aspect, in a possible implementation, the processing result generating unit further includes: the color processing of the indicator light of the image to be tested includes: using the HSV color model to define the range of required colors.

[0033] In conjunction with the second aspect, in a possible implementation, the processing result generating unit further includes: the color processing of the indicator light of the image to be tested also includes: using color distance debugging;

[0034] The color distance debugging is to mark the comparison distance value of each indicator light on the indicator light of the image to be tested, and return the calculation result according to the comparison distance value.

[0035] In combination with the second unit, in a possible implementation, the processing result generating unit further includes: the confidence array includes:

[0036] When color processing is performed on the image to be tested, a confidence value is determined according to the color distance.

[0037] In a third aspect, an embodiment of the present invention provides a cabinet indicator light recognition device based on template matching, the device comprising a camera, a memory, and a processor;

[0038] The memory is used to store computer executable instructions and images captured by the camera;

[0039] The processor is used to execute the computer-executable instructions to implement the method described in the first aspect and various possible implementations of the first aspect.

[0040] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium, characterized in that the computer-readable storage medium stores executable instructions, and when a computer executes the executable instructions, it can implement the method described in the first aspect and various possible implementation methods of the first aspect.

[0041] One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages:

[0042] The embodiment of the present invention adopts the method of obtaining a template image, obtaining a monitoring area, and determining a shape template matching image; using an image processing tool to determine a template position image including indicator lights and digital tubes for the shape template matching image; sorting the template position image in order to determine a position indication image; obtaining a picture to be tested, and determining a picture of the area to be tested; comparing the picture of the area to be tested with the position indication image to determine whether they match; if the matching result is yes, color processing is performed on the indicator lights of the picture to be tested, and a result array and a confidence array of the picture of the indicator lights to be tested are determined; and determining a result image according to the result array, which effectively solves the problems of false detection and missed detection in the prior art when manually inspecting cabinets, and inadequate inspection of cabinets at high positions or in harsh environments. And the problem of high cost is solved, and the method can be applied to the monitoring of indicator lights of computer room cabinets, so that manual participation is reduced. The device in the application is installed on the inspection robot, and the indicator lights in the cabinet are photographed, which greatly reduces the degree of manual participation, so that when inspecting high-risk cabinets with harsh environments, inspections can be carried out at a normal frequency, and after obtaining the result graph, a confidence array can be obtained, and the credibility of the method's judgment on the change of the indicator light can be obtained. The staff can judge whether the conclusion drawn by the method of the application is credible based on the credibility, and choose whether to manually re-check the cabinet indicator lights with low confidence, and there will be no problems of false detection or missed detection, and a large amount of labor costs are saved. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments of the present invention or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0044] Figure 1 A flowchart of the steps of the cabinet indicator light recognition method using template matching provided in an embodiment of the present application;

[0045] Figure 2 A shape template matching diagram of the cabinet indicator light recognition method using template matching provided in an embodiment of the present application;

[0046] Figure 3 The HSV model diagram in the cabinet indicator light recognition method based on template matching provided in the embodiment of the present application;

[0047] Figure 4 A schematic diagram of a cabinet indicator light recognition device for template matching provided in an embodiment of the present application;

[0048] Figure 5 A schematic diagram of a physical device of a cabinet indicator light identification device with template matching provided in an embodiment of the present application. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. The embodiments of the present invention and all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] In this embodiment, a method for identifying cabinet indicator lights by template matching is provided, and the method includes the following steps: Figure 1 As shown:

[0051] Step S101: Acquire a template image, obtain a monitoring area, and determine a shape template matching image.

[0052] Step S102: Acquire the template position map, sort the template position map in order, and determine the position indication map; the template position map is processed by an image processing tool on the shape template matching map to determine the digital map including indicator lights and digital tubes.

[0053] Step S103: Obtain the image to be tested and determine the image of the area to be tested.

[0054] Step S104: Compare the image of the area to be tested with the digital image to determine whether they match.

[0055] Step S105: If the matching result is yes, the indicator light of the image to be tested is subjected to color processing to determine the image result array and confidence array of the indicator light to be tested.

[0056] Step S106: Determine the result graph according to the result array.

[0057] The method effectively solves the problems of false detection and missed detection, inadequate inspection of cabinets at high positions or in harsh environments, and high cost in the prior art when inspecting cabinets manually, thereby realizing that the method can be applied to the monitoring of indicator lights of cabinets in computer rooms, thereby reducing manual involvement. The device in the present application is installed on an inspection robot, and photos of the indicator lights in the cabinets are taken, which greatly reduces the degree of manual involvement, so that when inspecting cabinets at high risks or in harsh environments, inspections can be carried out at a normal frequency, and after obtaining the result graph, a confidence array can be obtained, and the credibility of the method's judgment on indicator light changes can be obtained. The staff can judge whether the conclusions drawn by the method of the present application are credible based on the credibility, and choose whether to manually re-inspect the cabinet indicator lights with low confidence, and there will be no false detection and missed detection problems, and a large amount of labor costs are saved.

[0058] In step S101, a shape template matching map is determined, including: selecting four or more feature graphics on the template image, determining the area to be monitored, and the feature graphics are included in the template matching map. Since in actual operation, whether the template can be used for the cabinet indicator light to be identified depends on whether the area to be monitored can be completely overlapped, in step S101, four or more feature graphics are selected on the template map of the cabinet to be tested, and these feature graphics can frame the area to be tested in the template and find the area to be monitored in the image to be tested. Therefore, the role of the feature graphics is to find the unique feature mark of the area to be tested of the image to be tested, and it is also the unique identifier for framing the template area when generating the template, and the area determined by selecting four or more feature graphics of the template image of the cabinet to be tested is the shape surface model matching map. This step is mainly used to obtain the homography transformation matrix in the subsequent template matching process. The main idea of ​​the shape template matching adopted in this application is to traverse the pixel matrix calculation based on the overlap of the divergence.

[0059] In step S102, the template position diagram is sorted in order, including: using a scanning method from left to right and from top to bottom to sequentially label the indicator lights and digital tubes in the template position diagram. Preferably, in this application, Photoshop software or other common image processing software can be used to obtain the shape template matching diagram in step S101, and draw a template diagram containing the indicator lights and digital tubes, such as Figure 2 As shown, the indicator lights in the figure are scanned and numbered sequentially from left to right and from top to bottom to obtain a digital figure with numbers.

[0060] In step S105, the indicator light of the image to be tested is subjected to color processing, including: using the HSV color model to define the range of required colors. The HSV color model is as follows Figure 3As shown, the HSV color space is used to divide the color and brightness of the indicator lights in the test picture into intervals. In this application, the color of the cabinet indicator lights needs to be extracted as a reference for identification. There are many different types of cabinets in the computer room. The same color of the indicator lights on different cabinets has different properties such as color difference and saturation. When the staff performs manual identification, since people have innate prior knowledge learned, they can quickly identify that dark red or light red indicator lights are red. However, the program does not have this prior knowledge, so it needs to be added manually. The HSV model of color is used to extract the hue, saturation and brightness of all colors and save them in the file. In the identification stage, we need to divide the field of each identified knowledge light to be monitored to avoid adhesion between different lights, and then perform color discrimination on its central area. Specifically, the indicator lights that affect each other are divided into three standard colors of red, green and blue, and a color adjustment map is obtained.

[0061] In step S105, the indicator lights of the image to be tested are color processed, and the process also includes: using color distance debugging; using color distance debugging means marking the comparison distance value of each indicator light on the indicator light of the image to be tested, and returning the calculation result according to the comparison distance value. For debugging reference, the comparison distance value of each light on the enhanced test image and each light on the template image is marked. When the distance value is within the range of 0 to 100, it can be determined that its color has changed significantly compared with the template image. When the range is > 200, it is considered that its color value has not changed. When the given value is 100 to 200, the result to be determined will be returned.

[0062] In step S105, the confidence array includes: when the image to be tested is color processed, the confidence value is determined according to the color distance. The confidence value is directly related to the contrast distance value of the color distance. The larger the contrast distance value of the color distance, the confidence array indicates whether the color of the indicator light obtained in the HSV model color processing is reliable or not. The higher the confidence value, the higher the degree of trustworthiness. The lower the confidence value, the lower the degree of reliability. The staff can manually check the indicator lights with lower confidence values ​​according to the confidence value of the entire store obtained in the end, instead of manually checking each indicator light, which reduces the burden of manual work. In the entire monitoring process, there will be no missed detection, and the cabinet indicator lights can be inspected at a fixed frequency and fixed period in a computer room with a relatively poor working environment.

[0063] In the present application, the shape template matching map is first determined, and then four or more feature graphics in the captured image to be monitored are identified. The feature graphics mentioned here are consistent with the feature graphics in the shape template matching map. This part mainly uses the sliding window mode to scan alternate columns on the map. When all the preset number of graphics to be matched are found, the pixel coordinates of the image block are returned. Then, the image to be monitored is transformed into the field of view of the template map using the homography transformation. The homography transformation principle used in this process uses a camera to take images of the same plane object from different positions. There is a homography between them, which can be represented by a projection transformation. The projection transformation is a non-singular linear transformation under homogeneous coordinates. However, it is nonlinear in a non-homogeneous coordinate system, which shows that the invention of homogeneous coordinates is very valuable. The projection transformation has 2 more degrees of freedom than the affine transformation and has 8 degrees of freedom. The "invariant" property of the affine transformation mentioned above no longer exists in the projection transformation. Despite this, it still has an invariance, that is, the three points that remain collinear in the original image are still collinear after the transformation. The projection transformation is expressed as follows:

[0064]

[0065] Where V = (v 1 ,v 2 ) T , the scale of this vector changes with the position in the projection transformation, and the direction of the transformed line depends on the direction of the original line. Decomposing the projection transformation matrix into a chain of transformations, we get:

[0066]

[0067] Among them, H S is a similarity transformation, H A is an affine transformation, H P is a projective transformation, A=sRU+tV T , where U is an upper triangular matrix and after normalization det(U) = 1, v is obtained. If s is chosen to be an integer, then the decomposition is unique.

[0068] After obtaining the coordinates of more than four corresponding points, the Homography function in opencv can be used to calculate the homography transformation matrix H. After multiplying each homogeneous pixel coordinate of the image before transformation by H, the corresponding homogeneous pixel coordinate of the transformed image can be obtained. At this time, the part where the original image and the template image completely overlap can be taken out. According to the above principle, we only need to compare the digital image and the image to be monitored to obtain the result array. In this process, the aligned image to be tested is mainly binarized, and a series of operations such as corrosion and expansion are performed to obtain the connected domains of the indicator lights. Then, the indicator light map with flag bits can be obtained, and then each indicator light is color-measured. After determining its color, the result map can be obtained.

[0069] The present application also provides a cabinet indicator light recognition device with template matching, such as Figure 4 As shown, the device includes: a shape template determination unit 401, a template position determination unit 402, an input unit 403, an area matching unit 404, a processing result generation unit 405 and a result map generation unit 406. The shape template determination unit 401 is used to obtain a template image, obtain a monitoring area, and determine a shape template matching map. The template position determination unit 402 is used to obtain a template position map, and sort the template position map in order to determine a position indication map; the template position map is a digital map containing indicator lights and digital tubes determined after the shape template matching map is processed by an image processing tool. The input unit 403 is used to obtain a picture to be tested and determine a picture of the area to be tested. The area matching unit 404 is used to compare the picture of the area to be tested with the digital map to determine whether they match. The processing result generation unit 405 is used to determine that if the matching result is yes, the indicator lights of the picture to be tested are color processed to determine the result array and confidence array of the indicator light picture to be tested. The result map generation unit 406 is used to determine the result map according to the result array.

[0070] The above-mentioned device can effectively solve the problems of false detection and missed detection, inadequate inspection of cabinets at high positions or in harsh environments, and high cost in the prior art when manually inspecting cabinets, thereby realizing that the method can be applied to the indicator light monitoring of cabinets in computer rooms, so that manual participation is reduced. The camera 503 in the device of the present application is installed on the inspection robot to take pictures of the indicator lights in the cabinets, which greatly reduces the degree of manual participation, so that when inspecting cabinets at high risks or in harsh environments, they can be inspected at a normal frequency, and after obtaining the result graph, a confidence array can be obtained, and the credibility of the method's judgment on the indicator light change can be obtained. The staff can judge whether the conclusion drawn by the method of the present application is credible based on the credibility, and choose whether to manually re-check the cabinet indicator lights with low confidence, and there will be no false detection and missed detection problems, and a large amount of labor costs are saved.

[0071] In the device of the present application, the template position determination unit 401 is further used to select four or more characteristic patterns on the template image to determine the area to be monitored, and the characteristic patterns are included in the template matching image.

[0072] When the template position sorting unit 402 sorts the template position diagram in sequence, it uses a scanning method from left to right and from top to bottom to sequentially number the indicator lights and digital tubes in the template position diagram.

[0073] When performing color processing on the indicator light of the image to be tested, the processing result generating unit 405 uses the HSV color model to define the range of required colors.

[0074] The processing result generating unit 405 uses color distance debugging when performing color processing on the indicator lights of the tested image; using color distance debugging means marking the comparison distance value of each indicator light on the indicator lights of the tested image, and returning the calculation result according to the comparison distance value.

[0075] The confidence array includes: when color processing is performed on the image to be tested, the confidence value is determined according to the color distance.

[0076] like Figure 5 As shown, the present application also provides a cabinet indicator light recognition device based on template matching, which includes a camera 503, a memory 501, a processor 502, and a bus 504; the camera 503 is used to capture images and transmit the captured images to the processor 502 and the memory 501. The memory 501 is used to store computer executable instructions; the processor 502 is used to execute computer executable instructions to implement the cabinet indicator light recognition method based on template matching provided in this embodiment, and the bus 504 is used to transmit information.

[0077] An embodiment of the present invention further provides a computer-readable storage medium, and a computer can execute instructions to implement the cabinet indicator light recognition method based on template matching provided in this embodiment.

[0078] The above storage media include but are not limited to random access memory (RAM), read-only memory (ROM), cache, hard disk (HDD) or memory card. The memory can be used to store computer program instructions.

[0079] The devices or modules described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. For the convenience of description, the above devices are described in various modules according to their functions. When implementing this application, the functions of each module can be implemented in the same or multiple software and / or hardware. Of course, the module that implements a certain function can also be implemented by combining multiple sub-modules or sub-units.

[0080] Some modules in the apparatus described in the present application can be described in the general context of computer executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, classes, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0081] It can be known from the description of the above implementation mode that a person skilled in the art can clearly understand that the present application can be implemented by means of software plus necessary hardware. With such an understanding, the technical solution of the present application can be essentially or partly contributed to the prior art in the form of a software product, or it can be reflected in the implementation process of data migration. The computer software product can be stored in a storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes several instructions for a computer device (which can be a personal computer, a mobile terminal, a server, or a network device, etc.) to execute the methods described in the various embodiments of the present application or certain parts of the embodiments.

[0082] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. All or part of this application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld devices or portable devices, tablet devices, mobile communication terminals, multi-processor systems, microprocessor systems, programmable electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc.

[0083] The above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit the present application. Although the present application has been described in detail with reference to the aforementioned embodiments, a person of ordinary skill in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some or all of the technical features thereof may be replaced by equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the present application.

Claims

1. A cabinet indicator light recognition method based on template matching, characterized in that: include: Get the template image, get the monitoring area, and determine the shape template matching map; Acquire a template position map, and sort the template position map in order to determine a position indication map; the template position map is processed by an image processing tool on the shape template matching map to determine a digital map including indicator lights and digital tubes; Get the image to be tested and determine the image of the area to be tested; Compare the image of the area to be tested with the digital image to determine whether they match; If the matching result is yes, color processing is performed on the indicator light of the image to be tested, and a result array and a confidence array of the indicator light image to be tested are determined; Determine a result graph according to the result array; Selecting four or more characteristic patterns on the template image to determine the area to be monitored, and the characteristic patterns are included in the template matching image; The step of sorting the template position diagram in order comprises: using a scanning method from left to right and from top to bottom to sequentially label the indicator lights and digital tubes in the template position diagram to obtain a digital diagram with numbers; The color processing of the indicator light of the image to be tested includes: using the HSV color model to define the range of required colors; The color processing of the indicator light of the image to be tested also includes: using color distance debugging; The color distance debugging is to mark the comparison distance value of each indicator light on the indicator light of the image to be tested, and return the calculation result according to the comparison distance value; The confidence array includes: When color processing is performed on the image to be tested, a confidence value is determined according to the color distance.

2. A cabinet indicator light recognition device based on template matching, according to the cabinet indicator light recognition method based on template matching according to claim 1, characterized in that: include: Shape template determination unit: used to obtain the template image, obtain the monitoring area, and determine the shape template matching map; The template position determination unit is used to obtain a template position map, and sort the template position map in order to determine a position indication map; the template position map is processed by an image processing tool on the shape template matching map to determine a digital map including indicator lights and digital tubes; Input unit: used to obtain the image to be tested and determine the image of the area to be tested; Area matching unit: used to compare the image of the area to be tested with the digital image to determine whether they match; A processing result generating unit: used for judging that if the matching result is yes, then color processing is performed on the indicator light of the image to be tested, and determining the result array and confidence array of the image of the indicator light to be tested; A result graph generating unit is used to determine a result graph according to the result array.

3. A cabinet indicator light recognition device based on template matching, characterized in that: Includes camera, memory and processor; The camera is used to capture images and transmit the captured images to the processor and the memory; The memory is used to store computer executable instructions and images captured by the camera; The processor is configured to execute the computer executable instructions to implement the method of claim 1.

4. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores executable instructions, and when a computer executes the executable instructions, the method according to claim 1 can be implemented.

Images