A micro-track robot inspection method, track robot and system

Abstract

The present application relates to the field of rail robot inspection technology, and specifically to a micro rail robot inspection method, rail robot and system, including: the rail robot includes a walking track, a mobile frame, a lifting device, a driving wheel, a driving motor, a fixed frame, a battery, a binocular detection module, a communication transmission module and a motion control module, and the data of the hydropower station transmission line is obtained through the binocular detection module of the rail robot. The binocular detection module includes a thermal imager and a high-definition camera. Based on the local gray value distribution and difference of each edge pixel point, combined with the discrete degree and extreme difference of the local change value of the local edge pixel point, the gray fluctuation of each pixel point is determined; the heating fluctuation value is determined according to the local temperature change of the edge pixel point in the infrared image, and the damaged and heated area on the transmission line is detected. The rail robot transmits the detection result through the communication module. The present application can improve the inspection accuracy of the rail robot.

Description

Technical Field

[0001] The present application relates to the field of rail robot inspection technology, and in particular to a micro rail robot inspection method, a rail robot and a system. Background Art

[0002] With the development of technologies such as the Internet of Things, big data, and artificial intelligence, micro-rail robot inspection technology has been widely used. Compared with traditional manual inspections, micro-rail robot inspections have the advantages of being able to operate around the clock and being highly adaptable. Through pre-set inspection parameter information, micro-rail robots can automatically inspect and complete routine inspection tasks, and can issue intelligent alarms based on real-time inspection results, which can effectively improve inspection efficiency.

[0003] Micro-track robots are usually equipped with thermal imagers and high-definition cameras. They can conduct temperature and appearance inspections on the transmission lines in hydropower stations through the tracks built on the transmission lines. They can identify abnormal faults such as damage and heat on the transmission lines and provide intelligent early warnings.

[0004] However, in order to prevent the images taken in the micro-rail robot inspection environment from being affected by external interference factors, the image fusion method is often used to fuse the visible light image and the infrared image. However, the traditional method of using the maximum value to fuse the high-frequency components in the visible light image and the infrared image cannot accurately highlight the high-frequency characteristics of the damaged and heated transmission line, which can easily lead to blurred and low-quality damaged and heated areas on the transmission line in the fused image, making it impossible to accurately identify abnormal faults of damaged and heated transmission lines in hydropower stations, resulting in low accuracy of the micro-rail robot inspection and affecting the quality of the micro-rail robot inspection. Summary of the invention

[0005] In order to solve the above technical problems, the purpose of this application is to provide a micro-rail robot inspection method, a rail robot and a system. The technical solutions adopted are as follows:

[0006] The present application provides a micro-track robot inspection method, comprising the following steps:

[0007] The micro-track robot includes a walking track, a mobile frame, a lifting device, a driving wheel, a driving motor, a fixed frame, a battery, a binocular detection module, a communication transmission module and a motion control module. The binocular detection module of the micro-track robot is used to obtain an infrared image and a grayscale image of the transmission line in the hydropower station during the inspection process, wherein the binocular detection module includes a thermal imager and a high-definition camera;

[0008] Based on the information entropy of the grayscale values ​​of all edge pixels in the local window of each edge pixel in the grayscale image of the power transmission line and the grayscale difference between the edge pixels in the local window, the local change value of each edge pixel is obtained; wherein the local window is a preset window centered on the pixel;

[0009] The damage variation degree of each edge pixel point is obtained by the discrete degree and extreme difference of the local variation values ​​of all edge pixels points in the local window of each edge pixel point in the grayscale image of the power transmission line;

[0010] Analyze whether each pixel in the grayscale image of the transmission line belongs to an edge pixel, and determine the grayscale fluctuation of each pixel based on the local change value and damage change degree of the edge pixel;

[0011] According to the temperature change in the local window of the edge pixel points in the infrared image of the power transmission line, the heating fluctuation value of each pixel point is determined;

[0012] According to the gray value of each pixel in the high-frequency component image of the infrared image and the gray image of the transmission line, the gray fluctuation and the heating fluctuation value are combined to obtain a high-frequency fused image, and the low-frequency component image is fused, and the transmission line fused image is obtained by wavelet inverse transformation;

[0013] The threshold segmentation of the fused image of the transmission line is used to obtain the damaged and heated areas on the transmission line, and the micro-rail robot sends the inspection results through the communication transmission module.

[0014] Preferably, the local change value expression of each edge pixel point is:

[0015] , where is the local change value of edge pixel i in the grayscale image of the transmission line, is the information entropy of the grayscale values ​​of all edge pixels in the local window of edge pixel i, is the number of edge pixels in the local window of edge pixel i, and is the grayscale value of the tth and t-1th edge pixel in the local window of edge pixel i.

[0016] Preferably, the damage variation degree expression of each edge pixel point is: , where is the damage variation degree of edge pixel i in the grayscale image of the transmission line, are the coefficient of variation, maximum value and minimum value of the local change values ​​of all edge pixels in the local window of edge pixel i in the grayscale image of the transmission line.

[0017] Preferably, the grayscale fluctuation expression of each pixel point is: , where , , are the grayscale fluctuation, local change value and damage change degree of pixel i in the grayscale image of the transmission line respectively, and Q is the edge point set of the grayscale image of the transmission line.

[0018] Preferably, the heat fluctuation value of each pixel point is further:

[0019] For edge detection of infrared images of transmission lines, when a pixel does not belong to the edge pixel of the infrared image of the transmission line, the heat fluctuation value of the pixel is 0. Otherwise, the difference sequence of the local temperature sequence of each edge pixel is obtained, and the sum of all elements in the difference sequence is taken as the heat fluctuation value of each edge pixel.

[0020] Preferably, the local temperature sequence is a sequence composed of temperature values ​​of all pixels in a local window of edge pixels arranged from small to large temperatures.

[0021] Preferably, obtaining the high-frequency fused image further comprises:

[0022] High-frequency fusion value of pixel j in the high-frequency fusion image The expression is: , where is the infrared light fusion weight and gray value of pixel j in the high-frequency component image of the transmission line infrared image, is the visible light fusion weight and grayscale value of pixel j in the high-frequency component image of the transmission line grayscale image. The high-frequency fusion values ​​of all pixels constitute the high-frequency fusion image.

[0023] Preferably, the infrared light fusion weight and the visible light fusion weight of the pixel point j are respectively the normalized value of the heat fluctuation value of the pixel point corresponding to the pixel point j in the infrared image of the transmission line and the normalized value of the grayscale fluctuation value of the pixel point corresponding to the pixel point j in the grayscale image of the transmission line.

[0024] In addition, an embodiment of the present application provides a micro-rail robot, and the inspection of the rail robot is implemented using any of the micro-rail robot inspection methods described above.

[0025] An embodiment of the present application also provides a micro-rail robot inspection system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, wherein the processor implements the steps of any one of the above-mentioned methods when executing the computer program.

[0026] From the above, it can be seen that the micro-rail robot inspection method, rail robot and system provided by the present application have at least the following beneficial effects:

[0027] The present application is based on the analysis of the damaged detail features on the transmission line in the visible light image of the transmission line, and constructs the grayscale fluctuation degree, so that the damaged detail features need to be highlighted in the image fusion process, and the damaged position on the transmission line in the obtained fused image is avoided to be blurred. At the same time, based on the analysis of the damaged heating features in the infrared image of the transmission line, the heating fluctuation value is constructed, so that the subsequent fused image can highlight the features of the damaged heating area in the transmission line, thereby improving the accuracy of the micro-track robot inspection; and then the fusion weight is calculated according to the grayscale fluctuation degree and the heating fluctuation value, and the high-frequency component image of the infrared image of the transmission line and the high-frequency component image of the visible light image of the transmission line are fused with high-frequency components, and finally the fused image of the transmission line is obtained. Therefore, the present invention takes into account the damaged detail features and damaged heating features on the transmission line, and performs feature image fusion on the infrared image and visible light image of the transmission line captured by the thermal imager and the high-definition camera in the micro-track robot, so as to avoid the problems of blur and low quality in the infrared image and visible light image of the transmission line, and then accurately identify the abnormal faults of damaged heating on the transmission line in the hydropower station, and improve the accuracy of the micro-track robot inspection. BRIEF DESCRIPTION OF THE DRAWINGS

[0028] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present application or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0029] Figure 1 A flowchart of the steps of a micro-track robot inspection method provided in this application;

[0030] Figure 2 Grayscale image of the power transmission line inspected by the robot provided in this application;

[0031] Figure 3 Infrared images of power transmission lines inspected by robots provided for this application;

[0032] Figure 4 A schematic diagram of a high-frequency fusion image provided for this application;

[0033] Figure 5 Schematic diagram of low-frequency fusion image provided for this application;

[0034] Figure 6 The fused image of the power transmission line provided by this application;

[0035] Figure 7The fused image of the power transmission line obtained by the existing wavelet transform fusion algorithm provided in this application;

[0036] Figure 8 Schematic diagram of the construction modules of the micro-rail robot provided in this application. DETAILED DESCRIPTION

[0037] In order to further explain the technical means and effects adopted by the present application to achieve the predetermined invention purpose, the following, in combination with the accompanying drawings and preferred embodiments, describes in detail a micro-rail robot inspection method, rail robot and system proposed in the present application, its specific implementation method, structure, features and effects. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0038] Unless otherwise specified and limited, terms such as "comprises", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that a circuit structure, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such articles or devices. In the absence of further restrictions, an element defined by the sentence "comprising a ..." does not exclude the presence of other identical elements in the article or device including the element. In addition, the term "and\or" used herein includes any and all combinations of one or more related listed items. All technical and scientific terms used herein have the same meaning as those commonly understood by technicians in the technical field of this application.

[0039] The following is a detailed description of a micro-rail robot inspection method, a rail robot and a specific scheme of the system provided by the present application in conjunction with the accompanying drawings.

[0040] See also Figure 1 , which shows a flowchart of a micro-rail robot inspection method provided by an embodiment of the present application, including the following steps:

[0041] Step 1: Obtain infrared images and grayscale images of the transmission lines in the hydropower station during the inspection process of the micro-track robot.

[0042] The purpose of this embodiment is to avoid the problems of blur and low quality of infrared images and visible light images of transmission lines by performing feature image fusion on the infrared image obtained by the thermal imager in the micro-rail robot and the transmission line image (visible light image of transmission line) taken by the high-definition camera, and then accurately identify abnormal faults of damage and heat on the transmission line in the hydropower station, thereby improving the accuracy of the micro-rail robot's inspection.

[0043] In this embodiment, the micro-track robot includes a walking track, a mobile frame, a lifting device, a driving wheel, a driving motor, a fixed frame, a battery, a binocular detection module, a communication transmission module and a motion control module. The communication transmission module is used to transmit infrared images and visible light images of the transmission line; the motion control module is used to control the track robot to complete left drive, right drive and steering operations. It should be noted that the communication transmission module and the motion control module of the track robot adopt the communication transmission method and motion control method of the track robot in the prior art, and this embodiment does not impose special restrictions on this. The binocular detection module of the micro-track robot includes a thermal imager and a high-definition camera, which are used to collect infrared images and visible light images of the transmission line.

[0044] In order to improve the accuracy of the micro-rail robot inspection, the thermal imager and high-definition camera carried by the micro-rail robot are used to collect infrared images and visible light images of the transmission lines in the hydropower station during the inspection process in the same posture. During the image acquisition process, the size and viewing angle of the infrared image and the visible light image are kept consistent.

[0045] At the same time, in order to improve the accuracy of subsequent identification of abnormal damage and heating faults on transmission lines, the visible light image of the transmission line is converted into grayscale, and the visible light image of the transmission line in the hydropower station is used as the input of the grayscale average algorithm, and the output of the grayscale average algorithm is used as the grayscale image of the transmission line. The grayscale average algorithm is a well-known technology and the specific process will not be repeated here.

[0046] At this point, the infrared image and grayscale image of the transmission line can be obtained according to the above process of this embodiment. Specifically, the grayscale image and infrared image of the transmission line are respectively as follows: Figure 2 , 3 shown.

[0047] Step 2: Based on the information entropy of the grayscale values ​​of all edge pixels in the local window of each edge pixel in the grayscale image of the transmission line and the grayscale difference between the edge pixels in the local window, the local change value of each edge pixel is obtained; wherein the local window is a preset window centered on the pixel.

[0048] At present, the traditional method of fusing the high-frequency components in the visible light image and the infrared image using the maximum value cannot accurately highlight the high-frequency characteristics of the damaged and heated transmission line, which easily leads to the problem of blurring and low quality of the damaged and heated area on the transmission line in the fused image, and thus cannot accurately identify the abnormal faults of damaged and heated transmission lines in the hydropower station. Therefore, in order to avoid the problem of blurring and low quality of the damaged and heated area on the transmission line in the fused image, it is necessary to improve the fusion method of high-frequency components.

[0049] Generally speaking, the scale of the power system in a hydropower station is huge, and the structure of the transmission line in the hydropower station is highly complex. Since the temperature on the transmission line of a hydropower station is generally higher than the ambient temperature, the transmission line edge formed by the transmission line in the infrared image during the inspection of the micro-track robot is more obvious; at the same time, since there is a large difference between the edge formed by the transmission line and the background in the grayscale image of the transmission line, the transmission line in the hydropower station has a strong edge feature in the grayscale image of the transmission line.

[0050] In order to extract the edges formed by the transmission lines in the infrared image and the grayscale image of the transmission line, in this embodiment, the grayscale image and the infrared image of the transmission line are respectively used as inputs of the Sobel edge detection algorithm, and the Sobel edge detection algorithm is used to obtain edge pixel points on the grayscale image of the transmission line and edge pixel points on the infrared image of the transmission line, respectively. The Sobel edge detection algorithm is a well-known technology and the specific process will not be repeated here.

[0051] Under normal circumstances, the transmission lines in hydropower stations have grayscale standardized edge features. However, due to the influence of external environmental factors, the surface of the transmission lines is prone to damage and rust. The damage and rust on the transmission lines will cause the edges of the transmission lines to have irregular grayscale chaotic features. The larger the irregular grayscale chaotic features on the edges of the transmission lines, the more significant the damage and rust features are to a certain extent.

[0052] In order to analyze the edge features of damage and corrosion on the transmission line in the visible light image, preferably, in this embodiment, a window of size 15×15 is set with each edge pixel point in the grayscale image of the transmission line as the center, and is recorded as the local window of each edge pixel point; at the same time, the set composed of all edge pixels in the grayscale image of the transmission line is recorded as the edge point set Q of the grayscale image of the transmission line.

[0053] Based on the above analysis, the local change value of each edge pixel in the grayscale image of the transmission line is calculated based on the information entropy of the grayscale values ​​of all edge pixels in the local window of each edge pixel in the grayscale image of the transmission line and the grayscale difference between the edge pixels in the local window:

[0054] , where is the local change value of edge pixel i in the grayscale image of the transmission line, is the information entropy of the grayscale values ​​of all edge pixels in the local window of edge pixel i, is the number of edge pixels in the local window of edge pixel i, and is the grayscale value of the t-th and t-1-th edge pixel points in the local window of the edge pixel point i, wherein the calculation of information entropy is a well-known technology, and the specific process is not repeated here. The greater the information entropy of the grayscale value of the pixel points on the edge in the local area and the greater the grayscale difference between adjacent edge pixels, the more it can reflect the chaotic characteristics of the irregular grayscale on the edge of the transmission line, which to a certain extent indicates that the greater the characteristics of the damage and rust of the transmission line at the edge pixel point position, the greater the local change value.

[0055] Step 3: Obtain the damage variation degree of each edge pixel through the discrete degree and extreme value of the local variation values ​​of all edge pixels in the local window of each edge pixel in the grayscale image of the transmission line.

[0056] Generally speaking, the greater the change in the local change value between the edge pixels in the local area of ​​the transmission line grayscale image, the more it can reflect the difference in the change of the damage and rust characteristics between the pixels on the edge of the damaged area on the transmission line and the pixels on the normal edge of the transmission line, and the stronger the high-frequency detail characteristics of the damage and rust in the local area at the pixel position. When fusing the high-frequency components in the visible light image and infrared image of the transmission line, the high-frequency details of the damage and rust should be considered, so as to avoid the problem of blurring and low quality in the damaged and heated areas on the transmission line in the fused image, so as to accurately identify the abnormal faults of damage and heating on the transmission line in the hydropower station and improve the accuracy of the micro-track robot inspection.

[0057] Furthermore, in order to extract the high-frequency details of damage and corrosion in the visible light image of the transmission line, the damage change degree of each edge pixel in the grayscale image of the transmission line is calculated according to the distribution of the local change value of the edge pixel in the local window of each edge pixel:

[0058] , where is the damage variation degree of edge pixel i in the grayscale image of the transmission line, is the coefficient of variation of the local change values ​​of all edge pixels in the local window of edge pixel i in the grayscale image of the transmission line, and are the maximum and minimum values ​​of the local variation values ​​of all edge pixels in the local window of edge pixel i, respectively. The calculation of the coefficient of variation is a well-known technique, and the specific process is not repeated here. The larger the coefficient of variation and the larger the range, the more it can be explained that the damage and corrosion characteristics in the local area of ​​the transmission line have a larger variation feature. The more the high-frequency details of the damage and corrosion in the local area of ​​the transmission line can be reflected at this position, the greater the damage variation.

[0059] Step 4: Analyze whether each pixel in the grayscale image of the transmission line is an edge pixel, and determine the grayscale fluctuation of each pixel based on the local change value and damage change degree of the edge pixel.

[0060] Therefore, according to the above process of this embodiment, the damage and corrosion situation can be obtained. When the damage and corrosion characteristics appearing at the edge of the transmission line and the damage and corrosion change characteristics are large, the high-frequency details of the damage and corrosion at this position are stronger. In order to avoid the blurring and low quality of the damaged and heated areas on the transmission line in the fused image, it is necessary to extract the damage detail features in the grayscale image.

[0061] Based on the above analysis, it is analyzed whether each pixel in the grayscale image of the transmission line belongs to an edge pixel, and the grayscale fluctuation of each pixel in the grayscale image of the transmission line is calculated by combining the local change value and damage change degree of the edge pixel:

[0062] , where , , are the grayscale fluctuation, local change value, and damage change of pixel i in the grayscale image of the transmission line, respectively, and Q is the edge point set of the grayscale image of the transmission line. When the pixel does not belong to the edge pixel set of the grayscale image of the transmission line, the pixel position does not have the feature of reflecting the damage details on the edge of the transmission line, and the grayscale fluctuation is 0; when the pixel belongs to the edge pixel set of the grayscale image of the transmission line, the greater the damage and rust features and the changes in the damage and rust features, to a certain extent, the stronger the damage detail features of the transmission line at this position, the greater the grayscale fluctuation, and the damage detail features need to be highlighted in the image fusion process to avoid blurring of the damaged position on the transmission line in the fused image.

[0063] Step 5: Determine the heating fluctuation value of each pixel point based on the temperature change in the local window of the edge pixel point in the infrared image of the transmission line.

[0064] When damage features appear on the transmission line in a hydropower station, abnormal heating will occur at the damaged location of the transmission line. At the same time, there is a large difference between the temperature generated by the abnormal heating on the transmission line and the temperature at other normal locations on the transmission line, which often causes changes in temperature trend characteristics. For example, the temperature at the damaged location of the transmission line is at a higher level, while the normal temperatures on both sides of the damaged location on the transmission line are at a lower level. Since the temperature on the transmission line has a transfer effect, the temperature at the damaged location on the transmission line often shows a large trend change.

[0065] In order to identify the damage and heating characteristics in the infrared image of the transmission line, in this embodiment, a local window of 15×15 size is set with each pixel in the infrared image of the transmission line as the center. It should be noted that the pixel value of each pixel in the infrared image of the transmission line represents a temperature data.

[0066] In order to identify the temperature trend change around each pixel, for each pixel in the infrared image of the power transmission line, the sequence composed of the temperature values ​​of all pixels in the local window of each pixel is recorded as the local temperature sequence of each pixel. In this embodiment, the temperature values ​​of all pixels in the local window are arranged in order from small to large temperature to form a local temperature sequence.

[0067] Based on the above analysis, the heating fluctuation value of each pixel in the infrared image of the transmission line is calculated according to the temperature change in the local window of the edge pixel in the infrared image of the transmission line:

[0068] When a pixel point does not belong to an edge pixel point of the infrared image of the transmission line, the position of the pixel point does not have the characteristics of damage and heating on the edge of the transmission line, and the heating fluctuation value of the pixel point is 0;

[0069] When a pixel point belongs to an edge pixel point of the infrared image of the transmission line, the difference sequence of the local temperature sequence of each edge pixel point is calculated, and the sum of all elements in the difference sequence is recorded as the heat fluctuation value of the pixel point.

[0070] When a pixel point belongs to an edge pixel point in the infrared image of a transmission line, the greater the difference between adjacent temperature values ​​in the local temperature sequence, the greater the trend characteristic of the temperature rise around the pixel point position, and the more likely it is that the transmission line is damaged and heated, and the greater the heating fluctuation value.

[0071] Step 6: According to the grayscale value of each pixel in the high-frequency component image of the transmission line infrared image and the grayscale image, combined with the grayscale fluctuation and the heating fluctuation value, a high-frequency fused image is obtained, and the low-frequency component image is fused, and the transmission line fused image is obtained by inverse wavelet transform.

[0072] In order to fuse the high-frequency and low-frequency components of the visible light image and infrared image of the transmission line, the infrared image of the transmission line is used as the input of wavelet transform, and the high-frequency component image and low-frequency component image of the infrared image of the transmission line are obtained by wavelet transform.

[0073] Similarly, the grayscale image of the power transmission line is processed in the same way to obtain a high-frequency component image and a low-frequency component image of the grayscale image of the power transmission line.

[0074] Among them, wavelet transform is a well-known technology, and the specific process will not be repeated here.

[0075] Since the grayscale fluctuation of pixels in the grayscale image of the transmission line can reflect the high-frequency details of the damaged transmission line in the visible light image, and the heat fluctuation value of pixels in the infrared image of the transmission line can reflect the high-frequency details of the damaged and heated transmission line in the infrared image, setting the fusion weights during weighted fusion according to the grayscale fluctuation and the heat fluctuation value can highlight the damaged and heated areas in the transmission line, thereby avoiding the blurring and low quality problems of the damaged and heated areas on the transmission line in the fused image.

[0076] In order to realize high-frequency fusion of the high-frequency component images of the transmission line infrared image and the transmission line grayscale image, in this embodiment, the heat fluctuation values ​​of all pixels in the transmission line infrared image are z-score normalized, and the normalized value of the heat fluctuation value of each pixel in the transmission line infrared image is recorded as the infrared light fusion weight of the corresponding pixel in the high-frequency component image of the transmission line infrared image;

[0077] Similarly, in this embodiment, the z-score normalization method is used to normalize the grayscale fluctuation of all pixels in the grayscale image of the transmission line, and the normalized value of the grayscale fluctuation of each pixel in the grayscale image of the transmission line is recorded as the visible light fusion weight of each corresponding pixel in the high-frequency component image of the grayscale image of the transmission line.

[0078] Since both the infrared light fusion weight and the visible light fusion weight reflect the high-frequency characteristics of damage on the transmission line, the larger the infrared light fusion weight, the more likely it is that damage and heating will occur at the corresponding pixel point, and the clearer the high-frequency characteristics of damage and heating will be. When fusing high-frequency images, a larger weight should be given to highlight the significance of the damage and heating characteristics on the transmission line in the fused image. At the same time, the larger the visible light fusion weight, the more likely it is that damage will occur at the corresponding pixel point, and the clearer the high-frequency characteristics of the damage details at the pixel point, and the larger weight should be given to highlight the significance of the damage detail characteristics on the transmission line in the fused image.

[0079] Based on the above analysis, the high-frequency fusion value of each pixel in the fused high-frequency fusion image is calculated according to the grayscale value of each pixel in the high-frequency component image of the transmission line infrared image and the grayscale image, combined with the grayscale fluctuation and the heating fluctuation value.

[0080] In this embodiment, the high-frequency fusion value of pixel j in the high-frequency fusion image The expression is: , where is the infrared light fusion weight and gray value of pixel j in the high-frequency component image of the transmission line infrared image, is the visible light fusion weight and gray value of pixel j in the high-frequency component image of the grayscale image of the transmission line. The size of the infrared image and the grayscale image of the transmission line are consistent, and the pixels correspond one to one. The corresponding high-frequency component images are consistent in size and the pixels correspond one to one. Among them, the infrared light fusion weight and the visible light fusion weight are the normalized value of the heat fluctuation value of the corresponding pixel j in the infrared image of the transmission line and the normalized value of the grayscale fluctuation value of the corresponding pixel in the grayscale image of the transmission line, respectively.

[0081] It can be understood that the larger the infrared light fusion weight and the larger the visible light fusion weight, the more likely the pixel point is located at a damaged and heated location on the transmission line. The more significant the damaged and heated characteristics of the transmission line, the larger the high-frequency fusion value.

[0082] The image composed of the high-frequency fusion values ​​of all pixel points is recorded as the fused high-frequency fusion image. The high-frequency fusion value of the damaged and heated position of the transmission line in the high-frequency fusion image is larger, that is, the damaged and heated characteristics of the transmission line are more obvious.

[0083] Specifically, the schematic diagram of the high-frequency fusion image in this embodiment is as follows: Figure 4 shown.

[0084] Furthermore, the low-frequency components of the infrared image of the transmission line and the low-frequency components of the grayscale image of the transmission line are fused. The method of fusing the low-frequency components can be selected by the implementer in the actual application scenario. In this embodiment, the low-frequency component image of the infrared image of the transmission line and the low-frequency component image of the grayscale image of the transmission line are input by the average value method, and the low-frequency components in the infrared image of the transmission line and the grayscale image of the transmission line are fused by the traditional average value method to obtain a fused low-frequency fused image. The average value method is a well-known technology and the specific process will not be repeated here.

[0085] Specifically, the schematic diagram of the low-frequency fusion image in this embodiment is as follows: Figure 5 shown.

[0086] In order to realize the image fusion of the infrared image and the visible light image of the transmission line, the fused high-frequency fused image and the low-frequency fused image are used as wavelet inverse transform inputs, and the transmission line fused image is obtained by using wavelet inverse transform, wherein the wavelet inverse transform is a well-known technology and the specific process is not repeated here.

[0087] Specifically, in this embodiment, the fused transmission line fusion image is as follows: Figure 6 As shown in the figure, the transmission line fusion image obtained by the existing wavelet transform fusion algorithm is as follows: Figure 7As shown, compared with the transmission line fusion image obtained by the traditional wavelet transform fusion algorithm, the transmission line fusion image obtained in this embodiment has complete and clear details. In this embodiment, the grayscale fluctuation degree is constructed by analyzing the damage detail features on the transmission line in the visible light image of the transmission line, and the heat fluctuation value is constructed by analyzing the damage and heating features in the infrared image of the transmission line, so that the subsequent fused transmission line fusion image can highlight the features of the damaged and heated area in the transmission line, and avoid the phenomenon of discontinuity, blurring, etc. at the damaged position of the transmission line in the transmission line fusion image.

[0088] Step 7: Threshold segmentation of the fused image of the transmission line is performed to obtain the damaged and heated areas on the transmission line for fault inspection by the micro-rail robot.

[0089] Furthermore, considering that the features of the damaged and heated positions of the transmission lines in the fusion image of the transmission lines are more prominent, the threshold segmentation method is used to identify the abnormal faults of damaged and heated transmission lines in the hydropower station. The specific method is as follows:

[0090] The fused image of the transmission line is used as the input of the Otsu threshold segmentation algorithm, and the fused image area above the segmentation threshold is determined as the location area of ​​the damaged and heated transmission line. The Otsu threshold segmentation algorithm is used to obtain the damaged and heated area on the transmission line. The Otsu threshold segmentation algorithm is a well-known technology and the specific process will not be repeated here.

[0091] Furthermore, the micro-rail robot sends the detection results of damaged and heated areas on the transmission line to the inspection manager through the communication transmission module, and issues abnormal fault alarms to complete the inspection task of damaged and heated areas on the transmission line.

[0092] On the other hand, an embodiment of the present application provides a micro-rail robot, and the inspection of the rail robot is implemented by any one of the methods described in the above-mentioned micro-rail robot inspection method. The micro-rail robot includes a walking track, a mobile frame, a lifting device, a driving wheel, a driving motor, a fixed frame, a battery, a binocular detection module, a communication transmission module and a motion control module. The binocular detection module of the micro-rail robot obtains infrared images and grayscale images of the transmission lines in the hydropower station during the inspection, wherein the binocular detection module includes a thermal imager and a high-definition camera for collecting infrared images and visible light images of the transmission lines in the hydropower station; the communication transmission module is used to transmit the infrared images and visible light images of the transmission lines; the motion control module is used to control the rail robot to complete left drive, right drive and steering operations. It should be noted that the communication transmission module and the motion control module of the rail robot adopt the communication transmission method and motion control method of the rail robot in the prior art, and this embodiment does not impose any special restrictions on this.

[0093] Specifically, the schematic diagram of the construction module of the micro-track robot in this embodiment is as follows: Figure 8 shown.

[0094] Based on the same inventive concept as the above method, an embodiment of the present application also provides a micro-rail robot inspection system, including a memory, a processor, and a computer program stored in the memory and running on the processor, and when the processor executes the computer program, the steps of any one of the above-mentioned micro-rail robot inspection methods are implemented.

[0095] It is to be understood that the sequence of the embodiments of the present application described above is for description only and does not represent the advantages and disadvantages of the embodiments. The above describes specific embodiments of the present specification. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0096] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

[0097] The above content is only an implementation method of the present application and is not intended to limit the scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly used in other related technical fields, are also included in the protection scope of the present application.

Claims

1. A micro-track robot inspection method, characterized in that: The following steps are involved: The micro-track robot includes a walking track, a mobile frame, a lifting device, a driving wheel, a driving motor, a fixed frame, a battery, a binocular detection module, a communication transmission module and a motion control module. The binocular detection module of the micro-track robot is used to obtain an infrared image and a grayscale image of the transmission line in the hydropower station during the inspection process, wherein the binocular detection module includes a thermal imager and a high-definition camera; Based on the information entropy of the grayscale values ​​of all edge pixels in the local window of each edge pixel in the grayscale image of the power transmission line and the grayscale difference between the edge pixels in the local window, the local change value of each edge pixel is obtained; wherein the local window is a preset window centered on the pixel; The damage variation degree of each edge pixel point is obtained by the discrete degree and extreme difference of the local variation values ​​of all edge pixels points in the local window of each edge pixel point in the grayscale image of the power transmission line; Analyze whether each pixel in the grayscale image of the transmission line belongs to an edge pixel, and determine the grayscale fluctuation of each pixel based on the local change value and damage change degree of the edge pixel; According to the temperature change in the local window of the edge pixel points in the infrared image of the power transmission line, the heating fluctuation value of each pixel point is determined; According to the gray value of each pixel in the high-frequency component image of the transmission line infrared image and the gray image, combined with the gray fluctuation and the heating fluctuation value, a high-frequency fused image is obtained, and the low-frequency component image is fused, and the transmission line fused image is obtained by wavelet inverse transform; The threshold segmentation of the fused image of the transmission line is used to obtain the damaged and heated areas on the transmission line, and the micro-rail robot sends the inspection results through the communication transmission module.

2. A micro-track robot inspection method as claimed in claim 1, characterized in that: The local change value expression of each edge pixel point is: , where is the local change value of edge pixel i in the grayscale image of the transmission line, is the information entropy of the grayscale values ​​of all edge pixels in the local window of edge pixel i, is the number of edge pixels in the local window of edge pixel i, and is the grayscale value of the tth and t-1th edge pixel in the local window of edge pixel i.

3. A micro-track robot inspection method as claimed in claim 1, characterized in that: The damage variation expression of each edge pixel point is: , where is the damage variation degree of edge pixel i in the grayscale image of the transmission line, are the coefficient of variation, maximum value and minimum value of the local change values ​​of all edge pixels in the local window of edge pixel i in the grayscale image of the transmission line.

4. A micro-track robot inspection method as claimed in claim 1, characterized in that: The grayscale fluctuation expression of each pixel is: , where , , are the grayscale fluctuation, local change value and damage change degree of pixel i in the grayscale image of the transmission line respectively, and Q is the edge point set of the grayscale image of the transmission line.

5. A micro-track robot inspection method as claimed in claim 1, characterized in that: The heat fluctuation value of each pixel point is further: For edge detection of infrared images of transmission lines, when a pixel does not belong to the edge pixel of the infrared image of the transmission line, the heat fluctuation value of the pixel is 0. Otherwise, the difference sequence of the local temperature sequence of each edge pixel is obtained, and the sum of all elements in the difference sequence is taken as the heat fluctuation value of each edge pixel.

6. A micro-track robot inspection method as claimed in claim 5, characterized in that: The local temperature sequence is a sequence composed of temperature values ​​of all pixels in a local window of edge pixels arranged in ascending order of temperature.

7. A micro-track robot inspection method as claimed in claim 1, characterized in that: The obtaining of the high frequency fusion image further comprises: High-frequency fusion value of pixel j in the high-frequency fusion image The expression is: , where is the infrared light fusion weight and gray value of pixel j in the high-frequency component image of the transmission line infrared image, is the visible light fusion weight and grayscale value of pixel j in the high-frequency component image of the transmission line grayscale image. The high-frequency fusion values ​​of all pixels constitute the high-frequency fusion image.

8. A micro-track robot inspection method as claimed in claim 7, characterized in that: The infrared light fusion weight and visible light fusion weight of the pixel point j are respectively the normalized value of the heat fluctuation value of the pixel point corresponding to the pixel point j in the infrared image of the transmission line and the normalized value of the grayscale fluctuation value of the pixel point corresponding to the pixel point j in the grayscale image of the transmission line.

9. A micro-track robot, characterized in that: The inspection of the rail robot is achieved by using a micro rail robot inspection method as described in any one of claims 1-8.

10. A micro-track robot inspection system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 8 are implemented.

Images