Railway ballast state detection method and device based on point cloud difference graph

Abstract

The invention discloses a railway ballast state detection method and device based on a point cloud difference graph, and the method comprises the steps: carrying out the differential processing of actual elevation feature data and preset ideal elevation feature data, and obtaining a difference graph; obtaining a corresponding gray value region from the difference image, and determining the gray value region in a preset region range as a target region; constructing a target function according to a weight coefficient of a preset prior feature, and calculating a target function value by taking an area value under a target region, a Weber contrast ratio and an aspect ratio of the target region as constraint conditions; and comparing the target function value with a preset defect area effectiveness threshold value, and when the target function value is greater than the preset defect area effectiveness threshold value, determining that the target area is a defect area. According to the method, the positioning efficiency of the defect area can be improved, more accurate ballast bed detection in actual engineering is realized, and the detection cost is reduced.

Description

Technical Field

[0001] The present invention relates to the field of track detection technology, and in particular to a ballast state detection method and device based on a point cloud difference map. Background Art

[0002] This section is intended to provide a background or context to the embodiments of the invention recited in the claims. No admission is made that the description herein is prior art by inclusion in this section.

[0003] The railway ballast layer is the superstructure of the track structure, on which the superstructure of the railway, namely the tracks and sleepers, is laid. It plays a key role in maintaining the stability and safety performance of the railway track. During the long-term railway operation, the roadbed layer may have defects including overballast and underballast, which are caused by factors such as stone settlement, foundation deformation and climatic conditions. These defects may have a negative impact on the quality and safety of railway operations. Therefore, regular roadbed inspection and maintenance are crucial. However, the current roadbed inspection methods cannot efficiently and accurately locate defective areas, resulting in high costs for roadbed inspection and maintenance. Existing methods are gradually unable to meet the growing needs of railway transportation. Summary of the invention

[0004] The embodiment of the present invention provides a ballast state detection method based on a point cloud difference map, which is used to improve the positioning efficiency of defective areas, realize more accurate roadbed detection in actual engineering, and reduce detection costs. The method includes:

[0005] Select ballast particles at each detection point for scanning, obtain actual ballast particle point cloud data, and determine actual elevation feature data based on the actual ballast particle point cloud data;

[0006] Performing differential processing on the actual elevation feature data and the preset ideal elevation feature data to obtain a differential map;

[0007] Obtaining the corresponding grayscale value area from the difference image, and determining the grayscale value area within the preset area range as the target area;

[0008] The objective function is constructed according to the weight coefficient of the preset prior features, and the objective function value is calculated with the area value under the target area, the Weber contrast and the aspect ratio of the target area as constraint conditions; the preset prior features include the area value of the defect area, the aspect ratio of the defect area and the Weber contrast;

[0009] The objective function value is compared with a preset defect region validity threshold, and when the objective function value is greater than the preset defect region validity threshold, the target region is determined to be a defect region.

[0010] The embodiment of the present invention further provides a ballast state detection device based on a point cloud difference map, which is used to improve the positioning efficiency of defective areas, realize more accurate roadbed detection in actual engineering, and reduce detection costs. The device includes:

[0011] The actual elevation feature data acquisition module is used to select ballast particles at each detection point for scanning, obtain actual ballast particle point cloud data, and determine actual elevation feature data based on the actual ballast particle point cloud data;

[0012] The differential map acquisition module is used to perform differential processing on the actual elevation feature data and the preset ideal elevation feature data to obtain a differential map;

[0013] A target region determination module is used to obtain a corresponding gray value region from the difference image, and determine a gray value region within a preset region range as a target region;

[0014] An objective function value calculation module is used to construct an objective function according to the weight coefficient of the preset prior features, and calculate the objective function value with the area value under the target area, the Weber contrast and the aspect ratio of the target area as constraints; the preset prior features include the area value of the defect area, the aspect ratio of the defect area and the Weber contrast;

[0015] The defect area determination module is used to compare the target function value with a preset defect area validity threshold, and when the target function value is greater than the preset defect area validity threshold, determine that the target area is a defect area.

[0016] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned ballast state detection method based on point cloud difference map when executing the computer program.

[0017] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the ballast state detection method based on the point cloud difference map is implemented.

[0018] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the ballast state detection method based on the point cloud difference image is implemented.

[0019] In an embodiment of the present invention, ballast particles are selected at each detection point for scanning to obtain actual ballast particle point cloud data, and actual elevation feature data is determined based on the actual ballast particle point cloud data; the actual elevation feature data and the preset ideal elevation feature data are differentially processed to obtain a differential map; the corresponding gray value area is obtained from the differential map, and the gray value area within the preset area range is determined as the target area; the objective function is constructed according to the weight coefficient of the preset prior feature, and the objective function value is calculated with the area value under the target area, the Weber contrast and the aspect ratio of the target area as constraints; the preset prior features include the defect area value, the aspect ratio and the Weber contrast of the defect area; the objective function value is compared with the preset defect area validity threshold, and when the objective function value is greater than the preset defect area validity threshold, the target area is determined to be a defect area. In the above process, the embodiment of the present invention locates the target area through the differential map, calculates the objective function value based on the target area, and determines the target area corresponding to the objective function value greater than the preset defect area validity threshold as the defect area, thereby improving the positioning efficiency of the defect area and realizing more accurate roadbed detection in actual engineering to reduce detection costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0021] Figure 1 is a flow chart of a ballast state detection method based on a point cloud difference map in an embodiment of the present invention;

[0022] Figure 2 is the MLC curve in the embodiment of the present invention;

[0023] Figure 3 is a column grayscale distribution curve of the reverse difference image in an embodiment of the present invention;

[0024] Figure 4 Schematic diagram of a ballast state detection device based on a point cloud difference map in an embodiment of the present invention;

[0025] Figure 5 It is a specific schematic diagram of a ballast state detection device based on a point cloud difference image in an embodiment of the present invention. DETAILED DESCRIPTION

[0026] To make the purpose, technical solution and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the exemplary embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention.

[0027] Figure 1 : is a flow chart of a ballast state detection method based on a point cloud difference map in an embodiment of the present invention, the method comprising:

[0028] Step 101, selecting ballast particles at each detection point for scanning, obtaining actual ballast particle point cloud data, and determining actual elevation feature data according to the actual ballast particle point cloud data;

[0029] Step 102, performing differential processing on the actual elevation feature data and the preset ideal elevation feature data to obtain a differential map;

[0030] Step 103, obtaining a corresponding grayscale value area from the difference image, and determining a grayscale value area within a preset area range as a target area;

[0031] Step 104, constructing an objective function according to the weight coefficient of the preset prior features, and calculating the objective function value with the area value under the target area, the Weber contrast and the aspect ratio of the target area as constraints; the preset prior features include the area value of the defect area, the aspect ratio of the defect area and the Weber contrast;

[0032] Step 105 , comparing the objective function value with a preset defect region validity threshold, and when the objective function value is greater than the preset defect region validity threshold, determining the target region as a defect region.

[0033] Each step is described in detail below.

[0034] In step 101, ballast particles are selected at each detection point for scanning to obtain actual ballast particle point cloud data, and actual elevation feature data is determined based on the actual ballast particle point cloud data.

[0035] In a specific embodiment, the elevation feature data δ of each scan line width along the horizontal direction is extracted. i In the i-th scan line, the elevation characteristic coefficient δ i The calculation formula is as follows:

[0036]

[0037] Among them, E i is the average elevation value in the i-th scan line, B is the deviation, σ i is the standard deviation of the elevation within the i-th scan line.

[0038] In step 102, the actual elevation feature data and the preset ideal elevation feature data are subjected to differential processing to obtain a differential map.

[0039] In a specific embodiment, the conversion process of the difference map is as follows:

[0040]

[0041] x r = Mileage[x p ]

[0042]

[0043] Among them, z is the horizontal coordinate of the differential profile, x is the horizontal coordinate of the mileage direction (the direction of train travel), and y is the vertical coordinate of the elevation. p ,x p ) represents the pixel coordinates in the difference map, z p is the pixel horizontal coordinate of the difference image profile, x p is the pixel ordinate in the mileage direction in the difference image; (z r ,x r ) represents the actual coordinates in the point cloud, z r is the horizontal coordinate of the point cloud, x r is the vertical coordinate of the point cloud, D s represents the sampling distance, Mileage[i] represents the mileage value of the i-th frame of the lidar image, and W st Indicates the standard ballast layer width, The pixel coordinates in the difference map are (z p ,x p ), I(x p ,y p ) represents the gray value of the corresponding pixel, Δh max is the maximum elevation difference, Δh min is the minimum elevation difference.

[0044] Through differential projection of elevation values, the differential map describes the deviation information between the actual elevation feature data and the preset ideal elevation feature data, where the grayscale value represents the elevation distance between the corresponding ballast point and the standard template.

[0045] In step 103, a corresponding grayscale value region is obtained from the difference image, and a grayscale value region within a preset region range is determined as a target region.

[0046] In one embodiment, after obtaining the corresponding gray value area from the difference image, the method includes:

[0047] The Michelson contrast technique is used to enhance the gray value area in the difference image to obtain an enhanced gray value area;

[0048] The gray value area within the preset area range is determined as the target area, including:

[0049] The enhanced gray value area within the preset area range is determined as the target area.

[0050] In a specific embodiment, the ballast layer is rugged and rough due to the uneven size of the ballast stones, which results in a wide range of background grayscale values ​​in the difference image. This non-uniformity can make it challenging to accurately identify and analyze defective areas. A difference image enhancement method based on Michelson contrast (MLC) is used, which blurs the normal area of ​​the roadbed layer (including sleepers and rails) in the difference image and enhances the defective area, while eliminating the interference of the track and sleepers in the difference image. Michelson contrast can measure the contrast of periodic patterns, such as sinusoidal gratings, and its formula is given by the following formula:

[0051]

[0052] Where L max and L min Represent the maximum brightness value and the minimum brightness value respectively. By analyzing the stress characteristics of the ballast layer, it is found that the elevation distribution of the ballast has similarity along the track direction. This similarity is observable in the difference map, where pixels with the same grayscale value are often aligned in the same column, achieving enhancement within each pixel column of the difference map.

[0053] The difference image is divided into some sub-images, and the Michelson contrast MLC transform is performed in each sub-image. The MLC is given by the following formula:

[0054]

[0055] Where I(x,y) represents the grayscale value of the corresponding pixel, and p is the parameter to be determined.

[0056] In addition, this study improved the initial model by analyzing the characteristics of the MLC curve. Figure 2 is the MLC curve in the embodiment of the present invention, Figure 2 (a) is the MLC curve when p=20. Figure 2(b) is the MLC curve with p=200. The smaller the p value is, the more the features in the high gray value area can be weakened and the features in the low gray value area can be enhanced. Therefore, in the enhancement process, the difference map is inverted, and then the background part with high gray value will be blurred, and the defect part with low gray value will be enhanced. In addition, in order to eliminate the interference from the rails and sleepers during detection, the gray value range of the sleepers and rails must be controlled to the right of the p parameter value.

[0057] Figure 3 is a column grayscale distribution curve of the reverse difference image in an embodiment of the present invention, Figure 3 The curve of column 205 in (a) includes the defective area but excludes the rails and sleepers. Figure 3 The curve of column 270 in (b) includes the sleeper area and the defect area, but does not include the track area. Figure 3 The curve with 330 columns in (c) belongs to the track area. When the sub-image passes through the middle area, the curve shows a periodic distribution of local minimum values, which correspond to the sleeper area. At this time, the p parameter value is the local minimum. When the sub-image passes through the defect-free background area, the curve will show an almost random distribution, and the p parameter value is the global minimum. When the sub-image passes through the defective background area, the curve will show an extreme valley value. The p parameter value is above the extreme valley value and is close to the general valley value.

[0058] Based on the above analysis, we can conclude that:

[0059] p=μ w -σ w

[0060] where μ w represents the average gray value of the sub-image, σ w represents the standard deviation of the subgraph, then MLC is obtained by the following formula:

[0061]

[0062] Finally, since only the low gray value area of ​​the difference image needs to be paid attention to, in order to further enhance the characteristics of the defect area, the part greater than 0 in MLC is truncated, resulting in:

[0063]

[0064] In step 104, an objective function is constructed according to the weight coefficient of the preset prior features, and the objective function value is calculated with the area value under the target area, the Weber contrast and the aspect ratio of the target area as constraints; the preset prior features include the area value of the defect area, the aspect ratio of the defect area and the Weber contrast.

[0065] In one embodiment, constructing an objective function according to a weight coefficient of a preset priori feature includes:

[0066] The optimization iterative algorithm is used to optimize the weight coefficient of the preset prior feature to obtain the optimal weight coefficient;

[0067] According to the optimal weight coefficient, the objective function is constructed.

[0068] In a specific embodiment, the area corresponding to the bottom 1% of the grayscale value is extracted as the target area. Next, it is necessary to locate the defect area and determine its validity. The validity of the area is determined based on preset prior features, which include Weber contrast, defect area value, and aspect ratio of the defect area. On the one hand, if the area or aspect ratio of the area is too small, it may be caused by noise or system error, and it will be filtered out. On the other hand, in order to reduce the false detection rate, the validity of the defect area is further verified by Weber contrast. The Weber contrast formula is as follows:

[0069]

[0070] Where f represents the average gray value of the defect area, and b represents the average gray value of the background.

[0071] Given a region:

[0072] Zone[x,y,w,h]

[0073] Among them, x represents the lower boundary coordinate in the horizontal direction, y represents the lower edge coordinate in the vertical direction, w represents the width of the area, and h represents the height of the area.

[0074] The objective function P(R) is constructed according to the weight coefficient of the preset prior features:

[0075]

[0076] Where w1, w2 and w3 represent the weight coefficients corresponding to the preset prior features. Area(R) represents the area of ​​the defect area. min represents the defect area threshold, AR(R) represents the aspect ratio of the defect area, AR min represents the aspect ratio threshold of the defect area, C(R) represents the Weber contrast of the defect area, and C min Represents the Weber contrast threshold of the defect area. Calculate the objective function value of the objective function P(R), which is used as the defect effectiveness coefficient of the given R area and the defect area effectiveness threshold T p For comparison, when P(R)≥T p When , the target area corresponding to P(R) is determined to be the defect area.

[0077] After the initial defect area is determined, repeated iterative detection is performed until no further valid defects are identified, thereby effectively reducing false positives and missed positives in the track bed detection.

[0078] In step 105, the objective function value is compared with a preset defect region validity threshold, and when the objective function value is greater than the preset defect region validity threshold, the target region is determined to be a defect region.

[0079] In one embodiment, after determining that the target area is a defect area, the method includes:

[0080] The excess ballast volume and the under-ballast volume of the defective area are calculated by using one of the cross-section method, the grid method and the triangular grid method.

[0081] In a specific embodiment, after locating the position of the ballast defect, it is necessary to calculate the excess or insufficient volume. Combined with the actual situation of the line, the following three schemes are specifically designed, namely, the cross-section method, the grid method, and the triangular grid method. These three methods have their own advantages and disadvantages in terms of accuracy and time consumption. The method can be selected and adjusted according to the needs in the project.

[0082] The present invention also provides a ballast state detection device based on a point cloud difference map, as described in the following embodiments. Since the principle of solving the problem by the device is similar to that of the ballast state detection method based on a point cloud difference map, the implementation of the device can refer to the implementation of the ballast state detection method based on a point cloud difference map, and the repeated parts will not be repeated. Figure 4 Schematic diagram of a ballast state detection device based on a point cloud difference map in an embodiment of the present invention, the device comprises:

[0083] The actual elevation feature data acquisition module 401 is used to select ballast particles at each detection point for scanning, obtain actual ballast particle point cloud data, and determine actual elevation feature data based on the actual ballast particle point cloud data;

[0084] The difference map acquisition module 402 is used to perform difference processing on the actual elevation feature data and the preset ideal elevation feature data to obtain a difference map;

[0085] The target region determination module 403 is used to obtain the corresponding gray value region from the difference image, and determine the gray value region within the preset region range as the target region;

[0086] The objective function value calculation module 404 is used to construct an objective function according to the weight coefficient of the preset prior features, and calculate the objective function value with the area value under the target area, the Weber contrast and the aspect ratio of the target area as constraints; the preset prior features include the area value of the defect area and the aspect ratio of the defect area;

[0087] The defect area determination module 405 is used to compare the target function value with a preset defect area validity threshold, and determine that the target area is a defect area when the target function value is greater than the preset defect area validity threshold.

[0088] In one embodiment, a gray value region enhancement module is further included, which is specifically used to:

[0089] The Michelson contrast technique is used to enhance the gray value area in the difference image to obtain an enhanced gray value area;

[0090] The target area determination module is specifically used for:

[0091] The enhanced gray value area within the preset area range is determined as the target area.

[0092] In one embodiment, the objective function value calculation module 404 is specifically used to:

[0093] The optimization iterative algorithm is used to optimize the weight coefficient of the preset prior feature to obtain the optimal weight coefficient;

[0094] According to the optimal weight coefficient, the objective function is constructed.

[0095] Figure 5 5 is a specific schematic diagram of a ballast state detection device based on a point cloud difference map in an embodiment of the present invention. In one embodiment, it also includes a defect area volume calculation module 501, which is specifically used to:

[0096] The excess ballast volume and the under-ballast volume of the defective area are calculated by using one of the cross-section method, the grid method and the triangular grid method.

[0097] An embodiment of the present invention also provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned ballast state detection method based on point cloud difference map when executing the computer program.

[0098] An embodiment of the present invention further provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the ballast state detection method based on the point cloud difference map is implemented.

[0099] An embodiment of the present invention further provides a computer program product, which includes a computer program. When the computer program is executed by a processor, the ballast state detection method based on the point cloud difference image is implemented.

[0100] In an embodiment of the present invention, ballast particles are selected at each detection point for scanning to obtain actual ballast particle point cloud data, and actual elevation feature data is determined based on the actual ballast particle point cloud data; the actual elevation feature data and the preset ideal elevation feature data are differentially processed to obtain a differential map; the corresponding gray value area is obtained from the differential map, and the gray value area within the preset area range is determined as the target area; the objective function is constructed according to the weight coefficient of the preset prior feature, and the objective function value is calculated with the area value under the target area, the Weber contrast and the aspect ratio of the target area as constraints; the preset prior features include the defect area value, the aspect ratio and the Weber contrast of the defect area; the objective function value is compared with the preset defect area validity threshold, and when the objective function value is greater than the preset defect area validity threshold, the target area is determined to be a defect area. In the above process, the embodiment of the present invention locates the target area through the differential map, calculates the objective function value based on the target area, and determines the target area corresponding to the objective function value greater than the preset defect area validity threshold as the defect area, thereby improving the positioning efficiency of the defect area and realizing more accurate roadbed detection in actual engineering to reduce detection costs.

[0101] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0102] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0103] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0104] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0105] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A ballast state detection method based on point cloud difference map, characterized in that: include: Select ballast particles at each detection point for scanning, obtain actual ballast particle point cloud data, and determine actual elevation feature data based on the actual ballast particle point cloud data; Performing differential processing on the actual elevation feature data and the preset ideal elevation feature data to obtain a differential map; Obtaining the corresponding grayscale value area from the difference image, and determining the grayscale value area within the preset area range as the target area; The objective function is constructed according to the weight coefficient of the preset prior features, and the objective function value is calculated with the area value under the target area, the Weber contrast and the aspect ratio of the target area as constraint conditions; the preset prior features include the area value of the defect area, the aspect ratio of the defect area and the Weber contrast; The objective function value is compared with a preset defect region validity threshold, and when the objective function value is greater than the preset defect region validity threshold, the target region is determined to be a defect region.

2. The method according to claim 1, characterized in that After obtaining the corresponding gray value area from the difference image, including: The Michelson contrast technique is used to enhance the gray value area in the difference image to obtain an enhanced gray value area; The gray value area within the preset area range is determined as the target area, including: The enhanced gray value area within the preset area range is determined as the target area.

3. The method according to claim 1, characterized in that The objective function is constructed based on the weight coefficients of the preset prior features, including: The optimization iterative algorithm is used to optimize the weight coefficient of the preset prior feature to obtain the optimal weight coefficient; According to the optimal weight coefficient, the objective function is constructed.

4. The method according to claim 1, characterized in that After determining the target area as a defect area, including: The excess ballast volume and the under-ballast volume of the defective area are calculated by using one of the cross-section method, the grid method and the triangular grid method.

5. A ballast state detection device based on point cloud difference map, characterized in that: include: The actual elevation feature data acquisition module is used to select ballast particles at each detection point for scanning, obtain actual ballast particle point cloud data, and determine actual elevation feature data based on the actual ballast particle point cloud data; The differential map acquisition module is used to perform differential processing on the actual elevation feature data and the preset ideal elevation feature data to obtain a differential map; A target region determination module is used to obtain a corresponding gray value region from the difference image, and determine a gray value region within a preset region range as a target region; An objective function value calculation module is used to construct an objective function according to the weight coefficient of the preset prior features, and calculate the objective function value with the area value under the target area, the Weber contrast and the aspect ratio of the target area as constraints; the preset prior features include the area value of the defect area, the aspect ratio of the defect area and the Weber contrast; The defect area determination module is used to compare the target function value with a preset defect area validity threshold, and when the target function value is greater than the preset defect area validity threshold, determine that the target area is a defect area.

6. The device according to claim 5, characterized in that It also includes a gray value area enhancement module, which is specifically used for: The Michelson contrast technique is used to enhance the gray value area in the difference image to obtain an enhanced gray value area; The target area determination module is specifically used for: The enhanced gray value area within the preset area range is determined as the target area.

7. The device according to claim 5, characterized in that Objective function value calculation module, specifically used for: The optimization iterative algorithm is used to optimize the weight coefficient of the preset prior feature to obtain the optimal weight coefficient; According to the optimal weight coefficient, the objective function is constructed.

8. The device according to claim 5, characterized in that It also includes a defect area volume calculation module, which is specifically used for: The excess ballast volume and the under-ballast volume of the defective area are calculated by using one of the cross-section method, the grid method and the triangular grid method.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 4 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

11. A computer program product, characterized in that The computer program product comprises a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 4 is implemented.

Images