Quantification system and method for railway ballast saturation of ballast track

Abstract

The invention discloses a ballast saturation quantification system and method, and relates to the technical field of ballast saturation detection, an image acquisition module is used for acquiring a grey-scale map and a depth map of a ballast bed, an image preprocessing module is used for preprocessing the grey-scale map, a sleeper detection module is used for extracting sleeper characteristics in the grey-scale map, and the depth map of the ballast bed is obtained. The real-time position of the sleeper in the ballast bed is obtained; the railway ballast measuring module is used for determining a steel rail area, a sleeper area and a railway ballast area by utilizing the depth map and the sleeper position, and further calculating to obtain the steel rail bottom height, the sleeper height, the actual value of the railway ballast height on the two sides of a ballast bed and the actual value of the railway ballast height in the center of the ballast bed; and the railway ballast saturation quantification module is used for quantifying the railway ballast saturation by taking the rail bottom height of the steel rail as a reference value of the railway ballast height at the two sides of the ballast bed and the sleeper height as a reference value of the railway ballast height at the center of the ballast bed and utilizing the two groups of reference values and actual values. The method can simply and efficiently detect the saturation degree of the ballast bed and the railway ballast, and forms a quantitative result.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ballast saturation detection, and in particular to a quantification system and method for ballast saturation of a ballasted track. BACKGROUND

[0002] Ballasted tracks are often used in the construction of common-speed heavy-load railway lines due to their low construction cost, small noise propagation range, and short construction period.

[0003] When a train is running, the wheels act on the rail, not only generating vertical and lateral forces, but also generating longitudinal forces due to wheel rolling, longitudinal sliding, and train braking forces, etc., so that the rail bears longitudinal forces under dynamic loads. When the longitudinal force is greater than the rail foundation resistance (including fastener resistance, joint resistance, and ballast resistance, etc.), the rail longitudinal displacement is caused. In summer, the rail joint is easily squeezed into a narrow gap, causing the rail to expand and run off the track. In winter, the rail joint is also expanded, causing damage or breakage of the rail, clamps, and bolts. The rail sleeper is also stretched, causing adverse effects on the track gauge and track direction, resulting in damage to the fasteners and sleepers. At the turnout, the point rail and the basic rail are not tightly attached, the point rail bounces or moves poorly, and even the interlocking device is affected. In order to prevent the rail from climbing and causing damage to the components of the ballast bed and to ensure the safety of train operation, the fundamental measure is to improve the longitudinal resistance of the track. In addition to replacing the failed anti-climbing equipment, the more important measure is to keep the ballast in the sleeper box full. Therefore, the ballast in the sleeper box needs to be regularly detected. At present, manual inspection is mostly used to judge the degree of ballast fullness depending on the experience of workers, which has low detection accuracy and cannot form a quantitative result. SUMMARY

[0004] Therefore, the present application provides a quantification system and method for ballast saturation of a ballasted track to solve the problems in the prior art.

[0005] In a first aspect, a quantification system for ballast saturation of a ballasted track is provided, comprising: In a possible implementation manner of the first aspect, the preprocessing of the collected grayscale image comprises: standardizing and scaling the grayscale image.

[0006] In a possible implementation manner of the first aspect, based on the collected depth map and the real-time position of the sleeper, the rail area, the sleeper area, and the ballast area are determined, comprising: Based on the real-time position of the sleeper, the sleeper coordinates are obtained and the corresponding sleeper area is generated, denoted as a first sleeper area; The collected depth map is taken as an original depth map, the original depth map is subjected to height threshold segmentation and morphological filtering based on the prior data of the rail position, and a segmented binary image containing the rail area is obtained. Based on the segmented binary image and the original depth image, a relative height value of the rail is obtained; Based on the relative height value of the rail and the actual height value of itself, a segmentation threshold value of the ballast in the original depth image is obtained and segmented to obtain a ballast region; The depth image corresponding to the ballast region is sequentially subjected to mean filtering and morphological dilation, and then is subtracted from the original depth image to obtain a difference image; The difference image is subjected to threshold segmentation to obtain an initial sleeper region; The depth image corresponding to the initial sleeper region is subjected to morphological filtering and AND operation with the first sleeper region to obtain the sleeper region; The ballast region is obtained by subtracting the ballast region from the sleeper region.

[0007] In a possible implementation manner of the first aspect, the rail bottom height is calculated by: The rail bottom height is calculated by: The rail surface height mean value of the left and right rails is calculated by: The standard height of the rail is calculated by:

[0008] In a possible implementation manner of the first aspect, the sleeper height is calculated by: The sleeper height is calculated by: The height mean value of the sleeper center is calculated by:

[0009] In a possible implementation manner of the first aspect, the ballast height actual value on both sides of the ballast is calculated by: The ballast height actual value on both sides of the ballast is calculated by: The height mean value of the ballast on both sides of the ballast is calculated by:

[0010] In a possible implementation manner of the first aspect, the ballast height actual value in the center of the ballast is calculated by: The ballast height actual value in the center of the ballast is calculated by: The height mean value of the ballast in the center of the ballast is calculated by:

[0011] In a possible implementation manner of the first aspect, the ballast saturation in the ballast is quantified by: The ballast height actual value on both sides of the ballast is calculated by: Reference values ​​of ballast height on both sides of the track bed and calculate the actual value of the ballast height in the center of the track bed Ballast height in the center of the track bed The saturation of ballast in the roadbed is quantified by the ratio of .

[0012] In a possible implementation manner of the first aspect, the preset position is a location where an electronic tag is pre-installed on the line.

[0013] A second aspect of the present application provides a method for quantifying ballast saturation of a ballasted track, comprising: Use the image acquisition equipment installed on the inspection vehicle to collect grayscale images and depth images of the roadbed at preset locations on the line; Preprocess the collected grayscale image; Use the preset target detection model to perform target detection on the pre-processed grayscale image to obtain the real-time position of the sleeper; Determine the rail area, sleeper area, and ballast area based on the collected depth map and the real-time position of the sleepers; calculate the rail bottom height based on the rails and the depth map; calculate the sleeper height based on the sleeper area and the depth map; calculate the actual ballast height on both sides of the track bed based on the rail area, the ballast area, and the depth map; calculate the actual ballast height in the center of the track bed based on the sleeper area, the ballast area, and the depth map; The rail bottom height is used as a reference value for the ballast height on both sides of the roadbed, and the sleeper height is used as a reference value for the ballast height in the center of the roadbed; the ballast saturation in the roadbed is quantified based on the reference value for the ballast height on both sides of the roadbed and the actual value for the ballast height on both sides of the roadbed, and the reference value for the ballast height in the center of the roadbed and the actual value for the ballast height in the center of the roadbed.

[0014] Its beneficial effects are as follows: the present invention discloses a quantification system and method for the saturation of ballasted track ballast, which uses an image acquisition device installed on a patrol vehicle to respectively acquire a grayscale image and a depth image of the roadbed at a preset position on the line, and then pre-processes the acquired grayscale image through an image pre-processing module. Then, the sleeper detection module obtains the real-time position of the sleeper in the roadbed by extracting the sleeper features in the pre-processed grayscale image; the ballast measurement module uses the depth image and the sleeper position to determine the rail area, The sleeper area and the ballast area are further calculated to obtain the rail bottom height and sleeper height, the actual values ​​of the ballast height on both sides of the roadbed, and the actual value of the ballast height in the center of the roadbed; finally, the ballast saturation quantification module uses the rail bottom height as the reference value for the ballast height on both sides of the roadbed and the sleeper height as the reference value for the ballast height in the center of the roadbed. The reference value and actual value of the ballast height on both sides of the roadbed and the reference value and actual value of the ballast height in the center of the roadbed are used to quantify the ballast saturation in the roadbed. The present invention can simply and efficiently detect the saturation of the roadbed ballast and generate quantitative results. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without any creative work.

[0016] Figure 1 Schematic diagram of a system for quantifying saturation of ballasted track provided in an embodiment of the present application; Figure 2 This is a workflow diagram of the image acquisition module provided in an embodiment of the present application; Figure 3 This is a workflow diagram of the image preprocessing module provided in an embodiment of the present application; Figure 4 This is a workflow diagram of the ballast measurement module provided in an embodiment of the present application; Figure 5 This is a flow chart of a method for quantifying the saturation of ballasted track ballast provided in an embodiment of the present application. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0018] In this application, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or apparatus. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus comprising the element.

[0019] Example 1 In the existing technology, in order to prevent rail creep from causing damage to various components of the trackbed, the fundamental measure to ensure the safety of train operation is to increase the longitudinal resistance of the track. In addition to replacing the failed anti-climbing equipment, the more important measure to increase the longitudinal resistance of the track is to keep the ballast in the sleeper box full. This requires regular inspection of the ballast in the sleeper box. Currently, most inspections are done manually, relying on the workers' experience to judge the fullness of the ballast. The inspection accuracy is low and it is impossible to form quantitative results.

[0020] Therefore, the present application provides a quantification system for ballasted track ballast saturation, such as Figure 1 Shown, including: Image acquisition module: uses the image acquisition device installed on the inspection vehicle to collect grayscale images and depth images of the roadbed at preset locations on the line; Image preprocessing module: preprocess the collected grayscale image; Sleeper detection module: Uses a preset target detection model to perform target detection on the pre-processed grayscale image to obtain the real-time position of the sleeper; Ballast measurement module: Based on the collected depth map and the real-time position of the sleepers, the rail area, sleeper area and ballast area are determined; based on the rails and the depth map, the rail bottom height is calculated; based on the sleeper area and the depth map, the sleeper height is calculated; based on the rail area, the ballast area and the depth map, the actual ballast height on both sides of the track bed is calculated; based on the sleeper area, the ballast area and the depth map, the actual ballast height in the center of the track bed is calculated; Ballast saturation quantification module: The rail bottom height is used as the reference value of the ballast height on both sides of the roadbed, and the sleeper height is used as the reference value of the ballast height in the center of the roadbed; the ballast saturation in the roadbed is quantified based on the ballast height reference value on both sides of the roadbed and the actual ballast height value on both sides of the roadbed, and the ballast height reference value in the center of the roadbed and the actual ballast height value in the center of the roadbed.

[0021] Please refer to Figure 2 , Figure 2 This is a workflow diagram for the image acquisition module provided in an embodiment of the present application. The image acquisition module consists of an image acquisition device installed on an inspection vehicle. The image acquisition device includes a light source, an industrial high-speed camera, and a fill light device. During operation, the light source and fill light device illuminate the roadbed, and the high-speed industrial camera captures a grayscale image and depth map (e.g., 3200 x 512 pixels) of the roadbed at a preset location on the route. The preset location is set by installing an electronic tag at a station along the route. When the inspection vehicle passes by the electronic tag, the high-speed camera is controlled to capture the grayscale image and depth map of the roadbed.

[0022] Please refer to Figure 3 , Figure 3 This is a workflow diagram of the image preprocessing module provided in an embodiment of the present application. The image preprocessing module is used to perform standardization and scaling on the collected grayscale image. The standardization is calculated according to the following formula: To normalize the input data, For the data after standardization, is the mean value of all data of ballasted trackbed, It is the standard deviation of all the data of the ballasted track bed; The grayscale image is normalized and then scaled to 1600*256 pixels.

[0023] Among them, the sleeper detection module performs target detection on the preprocessed grayscale image through a preset target detection model. The preset target detection model is such as the YOLO series or the Faster Rcnn series. This embodiment adopts YOLO5 in the YOLO series as the target detection model in the sleeper detection module, which consists of three parts: Backbone, Neck and Head. The Backbone part mainly extracts multi-scale features from the preprocessed trackbed image for use in subsequent detection tasks. The Neck part further fuses the extracted scale features to generate a multi-scale feature pyramid. The Head detection head mainly performs multi-scale sleeper detection on the feature pyramid. In YOLO5, CBL is a combination of Convolution, BatchNormalization, and LeakyRelu operations to complete a single feature extraction operation. Resunit is a residual unit composed of CBL, ensuring network depth. CSP1-X and CSP2-X are different feature extraction components composed of CBL and Resunit, capable of extracting more complex features. SPP is a pooling module composed of a cascade of MaxPools to generate features of different sizes. Fcous is composed of Slice and CBL. The object detection model extracts features from the sleepers and determines their real-time position and shape.

[0024] Please refer to Figure 4 , Figure 4 This is a workflow diagram of the ballast measurement module provided in an embodiment of the present application. The ballast measurement module combines the collected depth map and the extracted real-time position of the sleeper to complete the ballast measurement. Specifically, based on the prior data of the rail position, the depth map is segmented by a height threshold (the rail area is extracted) and morphological filtering is performed (opening operation removes isolated noise points, closing operation fills the gap between the rail fasteners to ensure regional connectivity) to obtain a segmented binary map containing the rail area; using the segmented binary map and the original depth map, the relative height value of the rail, that is, the height of the rail surface relative to the ground, is obtained; based on the relative height value of the rail and the actual height of the rail itself, the segmentation threshold is determined to segment the roadbed area in the original depth map, and the segmentation threshold is used to segment the roadbed area. The trackbed height is obtained by subtracting the actual rail height from the relative rail height. This is used as the segmentation threshold to segment the trackbed area. The depth map corresponding to the trackbed area is subjected to mean filtering (to smooth depth map noise and reduce interference from ballast particle fluctuations) and morphological dilation (to expand the trackbed area and compensate for missing edges in threshold segmentation). This is then subtracted from the original depth map to obtain a difference map. This difference map is then threshold-segmented to obtain the initial sleeper area. The sleeper area is highlighted through the above operations. Finally, the depth map corresponding to the initial sleeper area is morphologically filtered and logically ANDed with the area corresponding to the extracted real-time sleeper position coordinates to obtain the precise sleeper area. The precise ballast area is then obtained by performing a subtraction operation on the trackbed and sleeper areas.

[0025] The calculation of rail bottom height includes: is the height of the rail bottom, is the average height of the left and right rails, is the standard height of the rail; Calculation of sleeper height includes: is the sleeper height, is the average height of the center of the sleeper; Calculation of the actual ballast height on both sides of the track bed includes: is the actual value of the ballast height on both sides of the track bed, is the average height of the ballast on both sides of the track bed; Calculation of the actual value of the ballast height in the center of the track bed includes: is the actual value of the ballast height in the center of the track bed, It is the average height of the ballast in the center of the trackbed.

[0026] Among them, the ballast saturation quantification module uses the calculated rail bottom height as the reference value for the ballast height on both sides of the roadbed, and the sleeper height as the reference value for the ballast height in the center of the roadbed. It then calculates the ratio between the reference value of the ballast height on both sides of the roadbed and the actual value of the ballast height on both sides of the roadbed, and the ratio between the reference value of the ballast height in the center of the roadbed and the actual value of the ballast height in the center of the roadbed. The two ratios obtained by the above calculations can clearly quantify the saturation of the ballast in the areas on both sides and the center of the roadbed, providing decision-making suggestions for subsequent ballast corrections. This embodiment can simply and efficiently realize the detection of ballast in the roadbed and can form quantitative results. Compared with inspections relying on manual experience, it is more efficient and standardized.

[0027] In some embodiments, preprocessing the collected grayscale image includes: performing normalization and scaling on the grayscale image.

[0028] In some embodiments, determining the rail area, the sleeper area, and the ballast area based on the collected depth map and the real-time position of the sleeper includes: Based on the real-time position of the sleeper, the sleeper coordinates are obtained and a corresponding sleeper area is generated and recorded as a first sleeper area; The collected depth map is used as the original depth map, and based on the prior data of the rail position, the original depth map is segmented using a height threshold and morphologically filtered to obtain a segmented binary map containing the rail area; Based on the segmented binary image and the original depth image, obtaining a relative height value of the rail; Based on the relative height value and the actual height value of the rail, a segmentation threshold of the track bed in the original depth map is obtained and segmented to obtain a track bed area; The depth map corresponding to the roadbed area is subjected to mean filtering and morphological dilation in sequence, and then subtracted from the original depth map to obtain a difference map; Performing threshold segmentation on the difference image to obtain an initial area of ​​the sleeper; Performing morphological filtering on the depth map corresponding to the initial sleeper area, and performing an AND operation on the depth map and the first sleeper area to obtain the sleeper area; Subtracting the trackbed area from the sleeper area to obtain the ballast area.

[0029] In some embodiments, calculating the rail bottom height includes: is the height of the rail bottom, is the average height of the left and right rails, The standard height of the rail.

[0030] In some embodiments, calculating the sleeper height includes: is the sleeper height, It is the average height of the center of the sleeper.

[0031] In some embodiments, calculating the actual values ​​of the ballast heights on both sides of the track bed includes: is the actual value of the ballast height on both sides of the track bed, It is the average height of the ballast on both sides of the trackbed.

[0032] In some embodiments, calculating the actual value of the ballast height in the center of the track bed includes: is the actual value of the ballast height in the center of the track bed, It is the average height of the ballast in the center of the trackbed.

[0033] In some embodiments, quantifying saturation of ballast in the track bed includes: Calculate the actual value of the ballast height on both sides of the track bed Reference values ​​of ballast height on both sides of the track bed and calculate the actual value of the ballast height in the center of the track bed Ballast height in the center of the track bed The saturation of ballast in the roadbed is quantified by the ratio of .

[0034] In some embodiments, the preset location is a location where an electronic tag is pre-installed on the line.

[0035] Example 2 Based on the quantification system of ballasted track ballast saturation provided in the first embodiment of the present application, the second embodiment of the present application also provides a quantification method of ballasted track ballast saturation, such as Figure 5 Shown, including: Use the image acquisition equipment installed on the inspection vehicle to collect grayscale images and depth images of the roadbed at preset locations on the line; Preprocess the collected grayscale image; Use the preset target detection model to perform target detection on the pre-processed grayscale image to obtain the real-time position of the sleeper; Determine the rail area, sleeper area, and ballast area based on the collected depth map and the real-time position of the sleepers; calculate the rail bottom height based on the rails and the depth map; calculate the sleeper height based on the sleeper area and the depth map; calculate the actual ballast height on both sides of the track bed based on the rail area, the ballast area, and the depth map; calculate the actual ballast height in the center of the track bed based on the sleeper area, the ballast area, and the depth map; The rail bottom height is used as a reference value for the ballast height on both sides of the roadbed, and the sleeper height is used as a reference value for the ballast height in the center of the roadbed; the ballast saturation in the roadbed is quantified based on the reference value for the ballast height on both sides of the roadbed and the actual value for the ballast height on both sides of the roadbed, and the reference value for the ballast height in the center of the roadbed and the actual value for the ballast height in the center of the roadbed.

[0036] Professionals may further appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computing software, or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the above description has generally described the components and steps of each example according to their functions. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professionals and technicians may use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0037] Although the preferred embodiments of the present invention have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0038] Obviously, those skilled in the art may make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if such changes and modifications fall within the scope of the claims and their equivalents, the present invention is intended to include such changes and modifications.

Claims

1. A quantification system for ballast saturation of ballasted track, characterized in that: include: Image acquisition module: uses the image acquisition device installed on the inspection vehicle to collect grayscale images and depth images of the roadbed at preset locations on the line; Image preprocessing module: preprocess the collected grayscale image; Sleeper detection module: Uses a preset target detection model to perform target detection on the pre-processed grayscale image to obtain the real-time position of the sleeper; Ballast measurement module: Determines the rail area, sleeper area, and ballast area based on the collected depth map and the real-time position of the sleepers; calculates the rail bottom height based on the rails and the depth map; Calculating the sleeper height based on the sleeper area and the depth map; Based on the rail area, the ballast area and the depth map, the actual ballast height values ​​on both sides of the roadbed are calculated; based on the sleeper area, the ballast area and the depth map, the actual ballast height value in the center of the roadbed is calculated; Ballast saturation quantification module: The rail bottom height is used as the reference value of the ballast height on both sides of the roadbed, and the sleeper height is used as the reference value of the ballast height in the center of the roadbed; the ballast saturation in the roadbed is quantified based on the ballast height reference value on both sides of the roadbed and the actual ballast height value on both sides of the roadbed, and the ballast height reference value in the center of the roadbed and the actual ballast height value in the center of the roadbed.

2. The quantification system for ballast saturation of ballasted track according to claim 1, characterized in that: The preprocessing of the collected grayscale image includes: performing standardization processing and scaling processing on the grayscale image.

3. The quantification system for ballasted track ballast saturation according to claim 1, characterized in that: Based on the acquired depth map and the real-time position of the sleepers, the rail area, sleeper area and ballast area are determined, including: Based on the real-time position of the sleeper, the sleeper coordinates are obtained and a corresponding sleeper area is generated and recorded as a first sleeper area; The collected depth map is used as the original depth map, and based on the prior data of the rail position, the original depth map is segmented using a height threshold and morphologically filtered to obtain a segmented binary map containing the rail area; Based on the segmented binary image and the original depth image, obtaining a relative height value of the rail; Based on the relative height value and the actual height value of the rail, a segmentation threshold of the track bed in the original depth map is obtained and segmented to obtain a track bed area; The depth map corresponding to the roadbed area is subjected to mean filtering and morphological dilation in sequence, and then subtracted from the original depth map to obtain a difference map; Performing threshold segmentation on the difference image to obtain an initial area of ​​the sleeper; Performing morphological filtering on the depth map corresponding to the initial sleeper area, and performing an AND operation on the depth map and the first sleeper area to obtain the sleeper area; Subtracting the trackbed area from the sleeper area to obtain the ballast area.

4. The quantification system for ballasted track ballast saturation according to claim 1, characterized in that: Calculation of rail bottom height includes: is the height of the rail bottom, is the average height of the left and right rails, The standard height of the rail.

5. The quantification system for ballasted track ballast saturation according to claim 4, characterized in that: Calculation of sleeper height includes: is the sleeper height, It is the average height of the center of the sleeper.

6. The quantification system for ballasted track ballast saturation according to claim 5, characterized in that: Calculation of the actual ballast height on both sides of the track bed includes: is the actual value of the ballast height on both sides of the track bed, It is the average height of the ballast on both sides of the trackbed.

7. The quantification system for ballasted track ballast saturation according to claim 6, characterized in that: Calculation of the actual value of the ballast height in the center of the track bed includes: is the actual value of the ballast height in the center of the track bed, It is the average height of the ballast in the center of the trackbed.

8. The quantification system for ballasted track ballast saturation according to claim 7, characterized in that: Quantifying the saturation of ballast within the trackbed involves: Calculate the actual value of the ballast height on both sides of the track bed Reference values ​​of ballast height on both sides of the track bed and calculate the actual value of the ballast height in the center of the track bed Ballast height in the center of the track bed The saturation of ballast in the roadbed is quantified by the ratio of .

9. The quantification system for ballasted track ballast saturation according to claim 1, characterized in that: The preset position is the location where the electronic tag is pre-installed on the line.

10. A method for quantifying ballast saturation of ballasted track, characterized in that: include: Use the image acquisition equipment installed on the inspection vehicle to collect grayscale images and depth images of the roadbed at preset locations on the line; Preprocess the collected grayscale image; Use the preset target detection model to perform target detection on the pre-processed grayscale image to obtain the real-time position of the sleeper; Determine the rail area, sleeper area, and ballast area based on the collected depth map and the real-time position of the sleepers; calculate the rail bottom height based on the rails and the depth map; Calculating the sleeper height based on the sleeper area and the depth map; Based on the rail area, the ballast area and the depth map, the actual ballast height values ​​on both sides of the roadbed are calculated; based on the sleeper area, the ballast area and the depth map, the actual ballast height value in the center of the roadbed is calculated; The rail bottom height is used as a reference value for the ballast height on both sides of the roadbed, and the sleeper height is used as a reference value for the ballast height in the center of the roadbed; the ballast saturation in the roadbed is quantified based on the reference value for the ballast height on both sides of the roadbed and the actual value for the ballast height on both sides of the roadbed, and the reference value for the ballast height in the center of the roadbed and the actual value for the ballast height in the center of the roadbed.

Images