Pantograph slide plate detection method and device

Abstract

The present application provides a pantograph skateboard detection method and device, wherein the pantograph skateboard detection method comprises: based on a preset image acquisition camera group, collecting a first image of the pantograph skateboard of the target train, wherein the image acquisition camera group includes fill lights of different bands and cameras of different bands; cropping the first image, and performing image enhancement on the cropping result to obtain a second image; processing the second image according to a semantic segmentation algorithm, and determining the edge of the pantograph skateboard based on the processing result; and determining the wear condition of the pantograph skateboard according to the edge of the pantograph skateboard. Through the layout of the image acquisition camera group and the improvement of the image processing method, the accuracy and stability of the pantograph skateboard wear detection are improved, false alarms are avoided, and the accuracy of the detection is maintained under different environmental conditions such as rain and sunlight.

Description

Technical Field

[0001] The present application relates to the field of rail transit technology, and in particular to a pantograph slide plate detection method, a pantograph slide plate detection device, a computing device, and a computer-readable storage medium. Background Art

[0002] The pantograph is a key device for rail trains to obtain electricity from the contact network. Its performance status directly affects the safe and stable operation of the train. In the existing detection methods, trackside static or dynamic detection methods are usually used to determine the wear of the pantograph carbon plate. However, most detection schemes still rely on manual regular inspections or stop the vehicle for measurement under specific working conditions. This method is inefficient and difficult to reflect the actual wear condition of the pantograph in real time.

[0003] In addition, existing detection methods also include the use of optical imaging technology and image processing algorithms to achieve non-contact automatic detection of the pantograph carbon plate thickness. However, this method captures the surface image of the pantograph plate through a camera and combines it with a flash or other light source for auxiliary lighting to improve the shooting quality. However, due to factors such as shadows, ambient light interference, and exposure transitions, the measurement accuracy is limited.

[0004] Specifically, in the existing detection methods, the design of the fill light system does not fully consider the complex and changeable on-site environmental light conditions. For example, various ambient lights such as direct sunlight during the day and light reflections at night will affect the brightness distribution of the skateboard surface captured by the camera, resulting in reduced image contrast and affecting the recognition accuracy of the worn edge. The pantograph carbon skateboard usually has a layer of aluminum edging, which has a high reflectivity. The aluminum edging will produce strong specular reflection under the fill light, forming a highlight area, which obscures or confuses the image information of the actual wear area of ​​the skateboard, thereby reducing the detection accuracy. On rainy days or when there is water on the skateboard surface, the moisture will change the optical properties of the skateboard surface, producing scattering and reflection effects. After being illuminated by the fill light, the rain may cause the skateboard surface to present a blurred image, making it difficult for the image processing algorithm to accurately extract the wear features, increasing the possibility of false alarms. The pantograph skateboard may produce fragments, fall off or become uneven due to wear during use. These irregular parts absorb and reflect light differently from the normal skateboard surface, which may cause the fill light source to be unable to evenly illuminate the skateboard, causing the imaging quality of the local area to decline, thereby affecting the overall wear judgment.

[0005] Furthermore, in the existing detection methods, the left and right side fill lights directly or indirectly illuminate the skateboard area. Since no effective isolation measures are taken, the light emitted by the left and right light sources cross each other. When the left and right light sources are turned on at the same time, the light generated by each will overlap with each other, causing the skateboard surface to be partially too bright or too dark, forming high light spots, shadows and other undesirable imaging effects, making it difficult for the image processing algorithm to accurately extract the skateboard wear characteristics. In order to obtain the ideal fill light effect, the staff needs to spend a lot of time and energy to manually adjust the angle, position and power of the fill lights on both sides. Even so, it may not be possible to achieve the ideal uniform lighting state, thereby increasing the cost and complexity of equipment debugging and maintenance.

[0006] Therefore, a pantograph slide plate detection method is urgently needed to solve the problems existing in the existing detection methods, such as uneven slide plate imaging brightness, difficulty in debugging the fill light, image edge curve extraction is easily affected by contact network occlusion and block falling, and slide plate centerline offset detection is affected by the contact network pull-out value. Summary of the invention

[0007] In view of this, the embodiment of the present application provides a pantograph slide plate detection method to solve the technical defects existing in the prior art. The embodiment of the present application also provides a pantograph slide plate detection device, a computing device, and a computer-readable storage medium.

[0008] According to a first aspect of an embodiment of the present application, a pantograph slide plate detection method is provided, comprising:

[0009] Based on a preset image acquisition camera group, a first image of the pantograph plate of the target train is acquired, wherein the image acquisition camera group includes fill lights of different bands and cameras of different bands;

[0010] Cropping the first image, and performing image enhancement on the cropping result to obtain a second image;

[0011] Processing the second image according to a semantic segmentation algorithm, and determining an edge of the pantograph slide based on the processing result;

[0012] The wear condition of the pantograph slide plate is determined according to the edge of the pantograph slide plate.

[0013] Optionally, the collecting of the first image of the pantograph slide of the target train based on a preset image collection camera group includes:

[0014] When the target train triggers the first sensor in the image acquisition camera group, the rear side image of the pantograph slide plate is acquired by the storage wear camera group in the image acquisition camera group;

[0015] When the target train triggers the second sensor in the image acquisition camera group, the front and side images of the pantograph slide are acquired through the outbound wear camera group in the image acquisition camera group, wherein the rear side image and the front side image constitute the first image.

[0016] Optionally, the collecting of the rear side image of the pantograph slide plate by the storage wear camera group in the image collection camera group includes:

[0017] The rear side image of the pantograph slide plate is collected by the warehouse-entry left camera group and the warehouse-entry right camera group included in the warehouse-entry wear camera group, wherein the warehouse-entry left camera group includes a warehouse-entry left camera and a warehouse-entry right flash light arranged on the opposite side, and the warehouse-entry right camera group includes a warehouse-entry right camera and a warehouse-entry left flash light arranged on the opposite side, and the wavelength bands of the warehouse-entry left camera and the warehouse-entry right flash light are both the first wavelength band, and the wavelength bands of the warehouse-entry right camera and the warehouse-entry left flash light are both the second wavelength band, and the first wavelength band is different from the second wavelength band;

[0018] Correspondingly, the collecting of the front and side images of the pantograph slide plate by the outbound wear camera group in the image collecting camera group includes:

[0019] The front side image of the pantograph slide is collected by the outgoing left camera group and the outgoing right camera group included in the outgoing wear camera group, wherein the outgoing left camera group includes the outgoing left camera and the outgoing right flash light arranged on the opposite side, the outgoing right camera group includes the outgoing right camera and the outgoing left flash light arranged on the opposite side, and the bands of the outgoing left camera and the outgoing right flash light are both the third band, and the bands of the outgoing right camera and the outgoing left flash light are both the fourth band, and the third band is different from the fourth band.

[0020] Optionally, the cropping the first image and performing image enhancement on the cropping result to obtain the second image includes:

[0021] Based on a preset cropping size and cropping area, cropping the first image to obtain a target cropped image;

[0022] Based on a preset image processing algorithm, contrast enhancement and histogram equalization are performed on the target cropped image to obtain the second image.

[0023] Optionally, processing the second image according to a semantic segmentation algorithm, and determining an edge of the pantograph slide based on the processing result, includes:

[0024] Processing the second image according to a semantic segmentation algorithm to obtain a third image;

[0025] Separating the adhesion of valid pixels and invalid pixels in the third image by using a morphological image processing algorithm;

[0026] Filling holes in the separation result based on the flood filling method, and removing invalid pixels in the filling result, to obtain a fourth image;

[0027] The edge of the pantograph slide is determined based on the fourth image.

[0028] Optionally, determining the edge of the pantograph slide according to the fourth image includes:

[0029] Based on a bilinear interpolation method, the resolution of the fourth image is improved to obtain a fifth image;

[0030] Performing contour detection on the fifth image to determine the minimum circumscribed rectangle of the pantograph slide;

[0031] Determine a segmentation line according to the vertices of the minimum circumscribed rectangle, and segment the fifth image according to the segmentation line to obtain a sixth image including an upper edge of the pantograph slide and a seventh image including a lower edge of the pantograph slide;

[0032] The edge of the pantograph slide is determined according to the sixth image and the seventh image.

[0033] Optionally, determining the edge of the pantograph slide according to the sixth image and the seventh image includes:

[0034] Performing edge extraction on the second image to obtain a dense edge map;

[0035] Expanding the upper edge of the pantograph slide and the upper edge of the pantograph slide, and solving the intersection of the expansion result and the dense edge map to obtain a real edge area;

[0036] According to the fill light principle associated with the image acquisition camera group, pixels in the real edge area are removed to obtain an eighth image;

[0037] The eighth image is processed in sequence by grayscale variance filtering, gradient variance filtering, noise removal, segmentation edge completion algorithm, polynomial interpolation algorithm and smoothing algorithm to obtain the edge of the pantograph slide.

[0038] According to a second aspect of an embodiment of the present application, a pantograph slide plate detection device is provided, comprising:

[0039] The acquisition module is configured to acquire a first image of the pantograph plate of the target train based on a preset image acquisition camera group, wherein the image acquisition camera group includes fill lights of different bands and cameras of different bands;

[0040] A cropping module, configured to crop the first image and perform image enhancement on the cropping result to obtain a second image;

[0041] A segmentation module is configured to process the second image according to a semantic segmentation algorithm, and determine an edge of the pantograph slide based on the processing result;

[0042] The determination module is configured to determine the wear condition of the pantograph slide board according to the edge of the pantograph slide board.

[0043] According to a third aspect of an embodiment of the present application, a computing device is provided, including:

[0044] Memory and processor;

[0045] The memory is used to store computer executable instructions, and the processor implements the steps of the pantograph slide detection method when executing the computer executable instructions.

[0046] According to a fourth aspect of an embodiment of the present application, a computer-readable storage medium is provided, which stores computer-executable instructions, and when the instructions are executed by a processor, the steps of the pantograph slide detection method are implemented.

[0047] According to a fifth aspect of an embodiment of the present application, a chip is provided, which stores a computer program, and the computer program implements the steps of the pantograph slide detection method when executed by the chip.

[0048] The pantograph slide detection method provided in the present application is to collect the first image of the pantograph slide of the target train based on a preset image acquisition camera group, wherein the image acquisition camera group includes fill lights of different bands and cameras of different bands; crop the first image, and perform image enhancement on the cropping result to obtain a second image; process the second image according to the semantic segmentation algorithm, and determine the edge of the pantograph slide based on the processing result; and determine the wear of the pantograph slide according to the edge of the pantograph slide. By improving the system shooting layout, core imaging devices, and adopting key technical means such as deep learning algorithms, the accuracy and stability of pantograph slide wear detection can be improved, false alarms can be reduced, and different environmental conditions (such as rain, block drop interference, and sunlight interference) can be adapted to ensure that the detection indicators specified by industry standards can be met at high or low speeds, while simplifying the operating process and improving the reliability and service life of the equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be 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.

[0050] Figure 1 It is a flow chart of a pantograph slide plate detection method provided by an embodiment of the present application;

[0051] Figure 2 It is a schematic diagram of an image acquisition camera group of a pantograph slide plate detection method provided in an embodiment of the present application;

[0052] Figure 3 It is a schematic diagram of a camera group for storage wear of a pantograph slide plate detection method provided in an embodiment of the present application;

[0053] Figure 4 It is a schematic diagram of image processing of a pantograph slide plate detection method provided by an embodiment of the present application;

[0054] Figure 5 It is a structural schematic diagram of a pantograph slide plate detection device provided in one embodiment of the present application;

[0055] Figure 6 It is a structural block diagram of a computing device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0056] Many specific details are described in the following description to facilitate a full understanding of the present application. However, the present application can be implemented in many other ways than those described herein, and those skilled in the art can make similar generalizations without violating the connotation of the present application, so the present application is not limited by the specific implementation disclosed below.

[0057] The terms used in one or more embodiments of the present application are only for the purpose of describing specific embodiments, and are not intended to limit one or more embodiments of the present application. The singular forms of "a", "said" and "the" used in one or more embodiments of the present application and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used in one or more embodiments of the present application refers to and includes any or all possible combinations of one or more associated listed items.

[0058] It should be understood that, although the terms first, second, etc. may be used to describe various information in one or more embodiments of the present application, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of one or more embodiments of the present application, the first may also be referred to as the second, and similarly, the second may also be referred to as the first.

[0059] In the present application, a pantograph slide plate detection method is provided. The present application also relates to a pantograph slide plate detection device, a computing device, and a computer-readable storage medium, which are described in detail one by one in the following embodiments.

[0060] Figure 1 A flow chart of a pantograph slide plate detection method provided according to an embodiment of the present application is shown, which specifically includes the following steps:

[0061] Step S102: Based on a preset image acquisition camera group, a first image of the pantograph plate of the target train is acquired, wherein the image acquisition camera group includes fill lights of different bands and cameras of different bands;

[0062] Step S104: cropping the first image, and performing image enhancement on the cropping result to obtain a second image;

[0063] Step S106: Processing the second image according to a semantic segmentation algorithm, and determining the edge of the pantograph slide based on the processing result;

[0064] Step S108: Determine the wear condition of the pantograph slide according to the edge of the pantograph slide.

[0065] Furthermore, in step S102, based on the preset image acquisition camera group, the process of acquiring the first image of the pantograph slide of the target train is specifically implemented as follows in this embodiment:

[0066] When the target train triggers the first sensor in the image acquisition camera group, the rear side image of the pantograph slide plate is collected through the in-warehouse wear camera group in the image acquisition camera group; when the target train triggers the second sensor in the image acquisition camera group, the front side image of the pantograph slide plate is collected through the out-warehouse wear camera group in the image acquisition camera group, wherein the rear side image and the front side image constitute the first image.

[0067] Among them, Figure 2As shown in a schematic diagram of an image acquisition camera group for a pantograph skateboard detection method provided, sensor A1 is the first sensor, and sensor B1 is the second sensor. It should be noted that the first sensor and the second sensor are infrared sensors, photoelectric sensors, etc. The specific sensor type used is determined by the actual usage scenario and is not limited in this embodiment; the skateboard is the pantograph skateboard of the target train, and the image acquisition camera group is arranged in the detection area.

[0068] Based on this, when the target train enters the detection area and the pantograph skateboard of the target train passes the sensing position of the first sensor, the rear side image of the pantograph skateboard is collected by an inbound wear camera group including two inbound wear cameras and two flash lights. When the target train continues to travel and its pantograph skateboard passes the sensing position of the second sensor, the front side image of the pantograph skateboard is collected by an outbound wear camera group including two outbound wear cameras and two flash lights.

[0069] Furthermore, the process of collecting the rear side image of the pantograph slide plate by the storage wear camera group in the image collection camera group is specifically implemented as follows in this embodiment:

[0070] The rear side image of the pantograph slide plate is collected by the warehouse-entry left camera group and the warehouse-entry right camera group included in the warehouse-entry wear camera group, wherein the warehouse-entry left camera group includes a warehouse-entry left camera and a warehouse-entry right flash light arranged on the opposite side, and the warehouse-entry right camera group includes a warehouse-entry right camera and a warehouse-entry left flash light arranged on the opposite side, and the wavelength bands of the warehouse-entry left camera and the warehouse-entry right flash light are both the first wavelength band, and the wavelength bands of the warehouse-entry right camera and the warehouse-entry left flash light are both the second wavelength band, and the first wavelength band is different from the second wavelength band;

[0071] Among them, Figure 3 As shown in the schematic diagram of the camera group for entering the warehouse for detecting the pantograph slide plate, the light source 1-450 is the left flash light for entering the warehouse, the light source 1-650 is the right flash light for entering the warehouse, the camera 1-650 is the left camera for entering the warehouse, and the camera 1-450 is the right camera for entering the warehouse.

[0072] Based on this, the left flash for entering the warehouse is the fill light for the right camera for entering the warehouse, and the right flash for entering the warehouse is the fill light for the left camera for entering the warehouse, to ensure that the light is evenly irradiated on the skateboard and reflected to the camera; and the left camera for entering the warehouse and the right flash for entering the warehouse use the same band, that is, the first band, and the right camera for entering the warehouse and the left flash for entering the warehouse also use the same band, that is, the second band. The first band and the second band are not the same, such as Figure 3 As shown, light source 1-450 and camera 1-450 use 450nm, while light source 1-650 and camera 1-650 use 650nm.

[0073] Therefore, each slide plate is illuminated by two fill lights of different wavelengths on the side of the slide plate, and two high-resolution cameras of different wavelengths are responsible for shooting to ensure that the slide plate is not blocked by the contact network and high-quality image data is obtained. The image contains the full picture of the pantograph slide plate and its size information. After the image algorithm is processed in real time, the thickness curve of the pantograph slide plate can be obtained.

[0074] Correspondingly, the process of collecting the front and side images of the pantograph slide plate by the outbound wear camera group in the image collection camera group is specifically implemented as follows in this embodiment:

[0075] The front side image of the pantograph slide is collected by the outgoing left camera group and the outgoing right camera group included in the outgoing wear camera group, wherein the outgoing left camera group includes the outgoing left camera and the outgoing right flash light arranged on the opposite side, the outgoing right camera group includes the outgoing right camera and the outgoing left flash light arranged on the opposite side, and the bands of the outgoing left camera and the outgoing right flash light are both the third band, and the bands of the outgoing right camera and the outgoing left flash light are both the fourth band, and the third band is different from the fourth band.

[0076] Among them, the working mode of the outbound wear camera group is similar to that of the above-mentioned inbound wear camera group, which will not be elaborated here. It should be noted that in actual application scenarios, in order to reduce the data processing volume of subsequent image processing processes, the outbound wear camera group uses the same band as the above-mentioned inbound wear camera group, that is, the third band is equal to one of the first band or the second band, and the fourth band is equal to the other of the first band or the second band.

[0077] In summary, due to the eccentric wear of the thickness on both sides of the pantograph skateboard, it is necessary to detect the thickness on both sides of the pantograph skateboard to determine the minimum remaining wear of the pantograph skateboard. Therefore, by introducing a dual-band fill light and a high-resolution camera combination, each skateboard is supplemented by two fill lights of different bands to illuminate the side of the skateboard, and the two high-resolution cameras are responsible for shooting the side of the skateboard in their respective corresponding bands, which effectively shields interference from other light sources and prevents exposure transition, ensuring high-quality skateboard side image data and preventing mutual interference between the left and right light sources. The above layout and shooting method are helpful to more comprehensively and accurately detect the minimum remaining wear of the carbon skateboard.

[0078] Furthermore, in step S104, the process of cropping the first image and performing image enhancement on the cropping result to obtain the second image is specifically implemented as follows in this embodiment:

[0079] Based on a preset cropping size and cropping area, the first image is cropped to obtain a target cropped image; based on a preset image processing algorithm, contrast enhancement and histogram equalization are performed on the target cropped image to obtain the second image.

[0080] Here, taking the rear side image in the first image as an example, Figure 4 As shown in the image processing schematic diagram of a pantograph slide detection method provided, the rear side image is cropped according to the determined cropping size and cropping position. It should be noted that the cropping position can be determined by a pre-trained target recognition model, that is, the pantograph slide is identified by the target recognition model, and the position of the pantograph slide is located, and the position is used as the cropping position. The target cropped image of the acquisition result is then enhanced by contrast and histogram equalization, and the second image obtained is as shown in FIG. Figure 4 (a) shown.

[0081] Furthermore, in step S106, the second image is processed according to the semantic segmentation algorithm, and the edge of the pantograph slide is determined based on the processing result. In this embodiment, the specific implementation method is as follows:

[0082] According to the semantic segmentation algorithm, the second image is processed to obtain the third image; through the morphological image processing algorithm, the adhesion of valid pixels and invalid pixels in the third image is separated; based on the flood filling method, the holes in the separation result are filled, and the invalid pixels in the filling result are eliminated to obtain the fourth image; the edge of the pantograph slide is determined according to the fourth image.

[0083] Among them, following the above example, DeepLabv3+ is used to perform semantic segmentation training on the skateboard, and libtorch is used to output the weight file after tracing. The semantic segmentation image of the skateboard, i.e., the third image, is obtained by calling the interface of the model deployment module. Some noise in the segmented image is removed, and the adhesion part of valid pixels and invalid pixels is separated by morphology. The holes in the segmentation result are filled with the flood filling method, and finally a large range of invalid pixels are removed. The fourth image is as follows Figure 4 The image processing schematic diagram of a pantograph slide detection method provided Figure 4 (b) as shown.

[0084] Furthermore, the above process of determining the edge of the pantograph slide plate according to the fourth image is specifically implemented as follows in this embodiment:

[0085] Based on the bilinear interpolation method, the resolution of the fourth image is improved to obtain the fifth image; the contour detection is performed on the fifth image to determine the minimum circumscribed rectangle of the pantograph slide; according to the vertices of the minimum circumscribed rectangle, a dividing line is determined, and the fifth image is divided according to the dividing line to obtain a sixth image including the upper edge of the pantograph slide and a seventh image including the lower edge of the pantograph slide; according to the sixth image and the seventh image, the edge of the pantograph slide is determined.

[0086] The resolution of the fourth image is improved by bilinear interpolation. After obtaining the fifth image, the contour of the fifth image is detected, and then the minimum circumscribed rectangle of the skateboard contour is calculated. The four endpoints of the upper and lower edges of the pantograph skateboard are determined according to the four vertices of the minimum circumscribed rectangle. The midpoints of the two side endpoints are the endpoints of the dividing line. Figure 4 The image processing schematic diagram of a pantograph slide detection method provided Figure 4 (c) as shown.

[0087] Based on this, edge detection is performed on the segmented sixth image and the seventh image to obtain their upper and lower edges, the distances from the upper edge pixel points and the lower edge pixel points to the segmentation line are calculated, the upper edge line and the lower edge line are separated, and the edge of the pantograph slide is determined, where the upper edge line and the lower edge line are as follows: Figure 4 The image processing schematic diagram of a pantograph slide detection method provided Figure 4 (d) as shown.

[0088] Furthermore, the process of determining the edge of the pantograph slide plate according to the sixth image and the seventh image is specifically implemented as follows in this embodiment:

[0089] Perform edge extraction on the second image to obtain a dense edge map; expand the upper edge of the pantograph slide and the upper edge of the pantograph slide, and solve the intersection of the expansion result and the dense edge map to obtain a real edge area; according to the fill light principle associated with the image acquisition camera group, eliminate the pixels in the real edge area to obtain an eighth image; through grayscale variance filtering, gradient variance filtering, noise removal, segmentation edge completion algorithm, polynomial interpolation algorithm and smoothing algorithm, the eighth image is processed in sequence to obtain the edge of the pantograph slide.

[0090] Among them, edge extraction is performed on the second image to obtain a dense edge map, such as Figure 4 The image processing schematic diagram of a pantograph slide detection method provided Figure 4 As shown in (e), the upper edge line and the lower edge line are expanded, and the intersection with the dense edge map is obtained to obtain the real edge area, and the pixels that do not meet the fill light principle are removed to obtain the eighth image. Taking the upper edge as an example, grayscale variance filtering is performed to remove pixels in the area where the grayscale change is not obvious, and the uppermost edge is obtained. The current edge and the segmented edge are superimposed, and gradient variance filtering is performed to remove pixels in the area where the gradient change is not obvious, such as contact network interference, to obtain the uppermost edge, remove noise pixels, use the segmented edge to complete the current edge, and obtain the uppermost edge, as shown in Figure 4 The image processing schematic diagram of a pantograph slide detection method provided Figure 4 As shown in (f), the edge point set is saved, polynomial interpolation is performed on it, and the curve is smoothed at the same time. The lower edge is obtained in the same way to obtain the edge of the pantograph slide.

[0091] In summary, the pantograph skateboard detection method in this embodiment improves the image analysis method and combines the image acquisition camera group with an optimized layout design, which greatly reduces the influence of factors such as ambient light, aluminum edging reflection, rain and falling blocks on the imaging quality, effectively improves the accuracy and stability of wear detection, and overcomes the false alarm and missed alarm phenomena caused by imaging quality problems that are common in traditional detection methods.

[0092] Secondly, the image acquisition camera group in this embodiment can cover the detection of different pantograph types by adjusting the position and angle of the camera and fill light. It can adapt to the different bow structures of single and double skateboards and the requirements of various skateboard spacings, greatly improving the versatility and practicality.

[0093] In addition, through the reasonable configuration of the flash and camera angles and spectrum, that is, each skateboard is supplemented by two fill lights of different bands for side lighting, and two high-resolution cameras of different bands are used to shoot the side of the skateboard in their respective corresponding bands, and the flash is located on the opposite side of the camera. For example, the left camera uses the right flash to fill in the light, and the right camera uses the left flash to fill in the light. This ensures that the light is evenly irradiated on the skateboard and reflected into the camera, reducing the mutual interference between the left and right light sources, avoiding the generation of overly bright or dark areas, and ensuring that high-quality image data can be obtained even under complex environmental conditions, thereby ensuring the accurate extraction of the skateboard edge curve and simplifying the on-site operation process.

[0094] Moreover, the left camera and the right flash are in the same band, and the right camera and the left flash are in the same band; the above two bands are isolated from each other, ensuring that each camera only receives the light from the corresponding flash, thereby avoiding mutual interference between light sources of different bands. It not only helps to improve the imaging quality, but also effectively solves the false alarm problem caused by ambient light, aluminum edging reflection, rain and falling blocks in traditional detection methods, thereby enhancing the adaptability and accuracy of the system.

[0095] Corresponding to the above method embodiment, the present application also provides a pantograph slide plate detection device embodiment, Figure 5 FIG. 1 is a schematic diagram showing the structure of a pantograph slide plate detection device provided by an embodiment of the present application. Figure 5 As shown, the device comprises:

[0096] The acquisition module 502 is configured to acquire a first image of the pantograph plate of the target train based on a preset image acquisition camera group, wherein the image acquisition camera group includes fill lights of different bands and cameras of different bands;

[0097] A cropping module 504 is configured to crop the first image and perform image enhancement on the cropping result to obtain a second image;

[0098] The segmentation module 506 is configured to process the second image according to a semantic segmentation algorithm, and determine the edge of the pantograph slide based on the processing result;

[0099] The determination module 508 is configured to determine the wear condition of the pantograph slide board according to the edge of the pantograph slide board.

[0100] In an optional embodiment, the acquisition module 502 is further configured to:

[0101] When the target train triggers the first sensor in the image acquisition camera group, the rear side image of the pantograph slide plate is collected through the in-warehouse wear camera group in the image acquisition camera group; when the target train triggers the second sensor in the image acquisition camera group, the front side image of the pantograph slide plate is collected through the out-warehouse wear camera group in the image acquisition camera group, wherein the rear side image and the front side image constitute the first image.

[0102] In an optional embodiment, the acquisition module 502 is further configured to:

[0103] The rear side image of the pantograph slide plate is collected by the warehouse-entry left camera group and the warehouse-entry right camera group included in the warehouse-entry wear camera group, wherein the warehouse-entry left camera group includes a warehouse-entry left camera and a warehouse-entry right flash light arranged on the opposite side, and the warehouse-entry right camera group includes a warehouse-entry right camera and a warehouse-entry left flash light arranged on the opposite side, and the wavelength bands of the warehouse-entry left camera and the warehouse-entry right flash light are both the first wavelength band, and the wavelength bands of the warehouse-entry right camera and the warehouse-entry left flash light are both the second wavelength band, and the first wavelength band is different from the second wavelength band;

[0104] In an optional embodiment, the acquisition module 502 is further configured to:

[0105] The front side image of the pantograph slide is collected by the outgoing left camera group and the outgoing right camera group included in the outgoing wear camera group, wherein the outgoing left camera group includes the outgoing left camera and the outgoing right flash light arranged on the opposite side, the outgoing right camera group includes the outgoing right camera and the outgoing left flash light arranged on the opposite side, and the bands of the outgoing left camera and the outgoing right flash light are both the third band, and the bands of the outgoing right camera and the outgoing left flash light are both the fourth band, and the third band is different from the fourth band.

[0106] In an optional embodiment, the cropping module 504 is further configured to:

[0107] Based on a preset cropping size and cropping area, the first image is cropped to obtain a target cropped image; based on a preset image processing algorithm, contrast enhancement and histogram equalization are performed on the target cropped image to obtain the second image.

[0108] In an optional embodiment, the segmentation module 506 is further configured to:

[0109] According to the semantic segmentation algorithm, the second image is processed to obtain the third image; through the morphological image processing algorithm, the adhesion of valid pixels and invalid pixels in the third image is separated; based on the flood filling method, the holes in the separation result are filled, and the invalid pixels in the filling result are eliminated to obtain the fourth image; the edge of the pantograph slide is determined according to the fourth image.

[0110] In an optional embodiment, the segmentation module 506 is further configured to:

[0111] Based on the bilinear interpolation method, the resolution of the fourth image is improved to obtain the fifth image; the contour detection is performed on the fifth image to determine the minimum circumscribed rectangle of the pantograph slide; according to the vertices of the minimum circumscribed rectangle, a dividing line is determined, and the fifth image is divided according to the dividing line to obtain a sixth image including the upper edge of the pantograph slide and a seventh image including the lower edge of the pantograph slide; according to the sixth image and the seventh image, the edge of the pantograph slide is determined.

[0112] In an optional embodiment, the segmentation module 506 is further configured to:

[0113] Perform edge extraction on the second image to obtain a dense edge map; expand the upper edge of the pantograph slide and the upper edge of the pantograph slide, and solve the intersection of the expansion result and the dense edge map to obtain a real edge area; according to the fill light principle associated with the image acquisition camera group, eliminate the pixels in the real edge area to obtain an eighth image; through grayscale variance filtering, gradient variance filtering, noise removal, segmentation edge completion algorithm, polynomial interpolation algorithm and smoothing algorithm, the eighth image is processed in sequence to obtain the edge of the pantograph slide.

[0114] The pantograph slide detection device provided in the present application collects the first image of the pantograph slide of the target train based on a preset image acquisition camera group, wherein the image acquisition camera group includes fill lights of different bands and cameras of different bands; crops the first image, and performs image enhancement on the cropping result to obtain a second image; processes the second image according to a semantic segmentation algorithm, and determines the edge of the pantograph slide based on the processing result; and determines the wear of the pantograph slide according to the edge of the pantograph slide. By improving the system shooting layout, core imaging devices, and adopting deep learning algorithms and other key technical means, the accuracy and stability of the pantograph slide wear detection can be improved, false alarms can be reduced, and different environmental conditions (such as rain, block drop interference, and sunlight interference) can be adapted to ensure that the detection indicators specified by the industry standards can be met at high or low speeds, while simplifying the operating process and improving the reliability and service life of the equipment.

[0115] The above is a schematic scheme of a pantograph slide detection device of the present embodiment. It should be noted that the technical scheme of the pantograph slide detection device and the technical scheme of the pantograph slide detection method mentioned above belong to the same concept. For the details not described in detail in the technical scheme of the pantograph slide detection device, please refer to the description of the technical scheme of the pantograph slide detection method mentioned above. In addition, each component in the device embodiment should be understood as a functional module that must be established to implement each step of the program flow or each step of the method, and each functional module is not an actual functional division or separation definition. The device claim defined by such a group of functional modules should be understood as a functional module architecture that mainly implements the solution through the computer program recorded in the specification, and should not be understood as a physical device that mainly implements the solution through hardware.

[0116] Figure 6 The block diagram of a computing device 600 according to an embodiment of the present application is shown. The components of the computing device 600 include but are not limited to a memory 610 and a processor 620. The processor 620 is connected to the memory 610 via a bus 630, and the database 650 is used to store data.

[0117] The computing device 600 also includes an access device 640 that enables the computing device 600 to communicate via one or more networks 660. Examples of these networks include a public switched telephone network (PSTN), a local area network (LAN), a wide area network (WAN), a personal area network (PAN), or a combination of communication networks such as the Internet. The access device 640 may include one or more of any type of network interface (e.g., a network interface card (NIC)) whether wired or wireless, such as an IEEE 802.11 wireless local area network (WLAN) wireless interface, a World Wide Interoperability for Microwave Access (Wi-MAX) interface, an Ethernet interface, a universal serial bus (USB) interface, a cellular network interface, a Bluetooth interface, a near field communication (NFC) interface, and the like.

[0118] In one embodiment of the present application, the above components of the computing device 600 and Figure 6 Other components not shown in the figure may also be connected to each other, for example, via a bus. It should be understood that Figure 6 The computing device structure block diagram shown is only for the purpose of illustration, and is not intended to limit the scope of the present application. Those skilled in the art may add or replace other components as needed.

[0119] The computing device 600 may be any type of stationary or mobile computing device, including a mobile computer or mobile computing device (e.g., a tablet computer, a personal digital assistant, a laptop computer, a notebook computer, a netbook, etc.), a mobile phone (e.g., a smart phone), a wearable computing device (e.g., a smart watch, smart glasses, etc.), or other types of mobile devices, or a stationary computing device such as a desktop computer or PC. The computing device 600 may also be a mobile or stationary server.

[0120] The processor 620 is used to execute computer executable instructions for each step of the pantograph slide detection method.

[0121] The above is a schematic scheme of a computing device of this embodiment. It should be noted that the technical scheme of the computing device and the technical scheme of the above-mentioned pantograph slide detection method belong to the same concept, and the details not described in detail in the technical scheme of the computing device can be referred to the description of the technical scheme of the above-mentioned pantograph slide detection method.

[0122] An embodiment of the present application also provides a computer-readable storage medium, which stores computer instructions, and when the instructions are executed by a processor, they are used to execute the steps of the pantograph slide detection method.

[0123] The above is a schematic scheme of a computer-readable storage medium of this embodiment. It should be noted that the technical scheme of the storage medium and the technical scheme of the pantograph slide detection method mentioned above belong to the same concept, and the details not described in detail in the technical scheme of the storage medium can be referred to the description of the technical scheme of the pantograph slide detection method mentioned above.

[0124] An embodiment of the present application further provides a chip storing a computer program, which implements the steps of the pantograph slide detection method when executed by the chip.

[0125] The above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the embodiments and still achieve the desired results. 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.

[0126] The computer instructions include computer program codes, which may be in source code form, object code form, executable files or some intermediate forms, etc. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0127] It should be noted that, for the above-mentioned method embodiments, for the sake of simplicity of description, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described action sequence, because according to the present application, certain steps can be performed in other sequences or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0128] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0129] The preferred embodiments of the present application disclosed above are only used to help explain the present application. The optional embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of the present application. The present application selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present application, so that those skilled in the art can understand and use the present application well. The present application is only limited by the claims and their full scope and equivalents.

Claims

1. A pantograph slide plate detection method, characterized in that: include: Based on a preset image acquisition camera group, a first image of the pantograph plate of the target train is acquired, wherein the image acquisition camera group includes fill lights of different bands and cameras of different bands; Cropping the first image, and performing image enhancement on the cropping result to obtain a second image; Processing the second image according to a semantic segmentation algorithm, and determining an edge of the pantograph slide based on the processing result; The wear condition of the pantograph slide plate is determined according to the edge of the pantograph slide plate.

2. The method according to claim 1, characterized in that The method of collecting a first image of a pantograph slide of a target train based on a preset image collection camera group includes: When the target train triggers the first sensor in the image acquisition camera group, the rear side image of the pantograph slide plate is acquired by the storage wear camera group in the image acquisition camera group; When the target train triggers the second sensor in the image acquisition camera group, the front and side images of the pantograph slide are acquired through the outbound wear camera group in the image acquisition camera group, wherein the rear side image and the front side image constitute the first image.

3. The method according to claim 2, characterized in that The collecting of the rear side image of the pantograph slide plate by the storage wear camera group in the image collecting camera group includes: The rear side image of the pantograph slide plate is collected by the warehouse-entry left camera group and the warehouse-entry right camera group included in the warehouse-entry wear camera group, wherein the warehouse-entry left camera group includes a warehouse-entry left camera and a warehouse-entry right flash light arranged on the opposite side, and the warehouse-entry right camera group includes a warehouse-entry right camera and a warehouse-entry left flash light arranged on the opposite side, and the wavelength bands of the warehouse-entry left camera and the warehouse-entry right flash light are both the first wavelength band, and the wavelength bands of the warehouse-entry right camera and the warehouse-entry left flash light are both the second wavelength band, and the first wavelength band is different from the second wavelength band; Correspondingly, the collecting of the front and side images of the pantograph slide plate by the outbound wear camera group in the image collecting camera group includes: The front side image of the pantograph slide is collected by the outgoing left camera group and the outgoing right camera group included in the outgoing wear camera group, wherein the outgoing left camera group includes the outgoing left camera and the outgoing right flash light arranged on the opposite side, the outgoing right camera group includes the outgoing right camera and the outgoing left flash light arranged on the opposite side, and the bands of the outgoing left camera and the outgoing right flash light are both the third band, and the bands of the outgoing right camera and the outgoing left flash light are both the fourth band, and the third band is different from the fourth band.

4. The method according to claim 1, characterized in that: The step of cropping the first image and performing image enhancement on the cropping result to obtain a second image includes: Based on a preset cropping size and cropping area, cropping the first image to obtain a target cropped image; Based on a preset image processing algorithm, contrast enhancement and histogram equalization are performed on the target cropped image to obtain the second image.

5. The method according to claim 1, characterized in that The processing of the second image according to the semantic segmentation algorithm and determining the edge of the pantograph slide based on the processing result includes: Processing the second image according to a semantic segmentation algorithm to obtain a third image; Separating the adhesion of valid pixels and invalid pixels in the third image by using a morphological image processing algorithm; Filling holes in the separation result based on the flood filling method, and removing invalid pixels in the filling result, to obtain a fourth image; The edge of the pantograph slide is determined based on the fourth image.

6. The method according to claim 5, characterized in that The step of determining the edge of the pantograph slide according to the fourth image comprises: Based on a bilinear interpolation method, the resolution of the fourth image is improved to obtain a fifth image; Performing contour detection on the fifth image to determine the minimum circumscribed rectangle of the pantograph slide; Determine a segmentation line according to the vertices of the minimum circumscribed rectangle, and segment the fifth image according to the segmentation line to obtain a sixth image including an upper edge of the pantograph slide and a seventh image including a lower edge of the pantograph slide; The edge of the pantograph slide is determined according to the sixth image and the seventh image.

7. The method according to claim 6, characterized in that The step of determining the edge of the pantograph slide plate according to the sixth image and the seventh image includes: Performing edge extraction on the second image to obtain a dense edge map; Expanding the upper edge of the pantograph slide and the upper edge of the pantograph slide, and solving the intersection of the expansion result and the dense edge map to obtain a real edge area; According to the fill light principle associated with the image acquisition camera group, pixels in the real edge area are removed to obtain an eighth image; The eighth image is processed in sequence by grayscale variance filtering, gradient variance filtering, noise removal, segmentation edge completion algorithm, polynomial interpolation algorithm and smoothing algorithm to obtain the edge of the pantograph slide.

8. A pantograph slide plate detection device, characterized in that: include: The acquisition module is configured to acquire a first image of the pantograph plate of the target train based on a preset image acquisition camera group, wherein the image acquisition camera group includes fill lights of different bands and cameras of different bands; A cropping module, configured to crop the first image and perform image enhancement on the cropping result to obtain a second image; A segmentation module is configured to process the second image according to a semantic segmentation algorithm, and determine an edge of the pantograph slide based on the processing result; The determination module is configured to determine the wear condition of the pantograph slide board according to the edge of the pantograph slide board.

9. A computing device, characterized in that include: Memory and processor; The memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the steps of the pantograph slide detection method described in any one of claims 1 to 7.

10. A computer-readable storage medium storing computer instructions, characterized in that: When the instruction is executed by the processor, the steps of the pantograph slide detection method described in any one of claims 1 to 7 are implemented.

Images