Method and device for evaluating crack state of double-block ballastless track

Abstract

The invention provides a double-block ballastless track crack state evaluation method and device, and relates to the technical field of track railways, and the method comprises the following steps: obtaining image information of a double-block ballastless track in real time; the image information of the double-block ballastless track is sent to a preset image processing module for target detection, and crack image information and crack size information of the double-block ballastless track are obtained; determining crack state evaluation index information based on the crack image information and the crack size information of the double-block ballastless track, wherein the crack state evaluation index information comprises the grade and the number of crack state evaluation indexes; the crack state evaluation index information is sent to a preset crack state evaluation module to be processed, evaluation index weight calculation and crack state evaluation are conducted in sequence through a preset calculation formula, and a state evaluation result of the double-block ballastless track is obtained. According to the invention, the accuracy and efficiency of ballastless track crack detection are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of railroads, and in particular to a method and device for evaluating crack status of a double-block ballastless track. Background Art

[0002] With the rapid development of high-speed railways in my country, ballastless track technology has gradually become the mainstream choice in the construction of high-speed railway lines. Among them, the double-block ballastless track is widely used due to its good stability and durability. However, the roadbed of the double-block ballastless track is a cast-in-place concrete structure, and the sleepers are prefabricated structures. Environmental factors (such as temperature, humidity, climate, etc.) during the construction process are likely to cause a significant increase in the risk of roadbed cracking. Especially in long-term operation, the roadbed slab of the double-block ballastless track often has eight-shaped cracks at the sleeper ends and through cracks between sleepers. Most of these cracks occur at the junction of the sleepers and the roadbed slab concrete. Over time, the cracks gradually expand and may form through cracks, seriously affecting the performance and safety of the track. Existing technologies mainly rely on subjective crack quantity and appearance evaluation methods, but due to the lack of unified quantitative standards, the evaluation results often have large differences, making it difficult to provide an effective basis for subsequent maintenance, and there is a lack of automated evaluation methods based on image processing.

[0003] Most of the existing crack detection methods for dual-block ballastless tracks remain at the manual observation stage, and make rough judgments by visually observing the number and distribution of cracks. This is inefficient and easily interfered by human factors. This technical limitation makes it impossible for existing crack detection methods to provide accurate references for the status assessment of dual-block ballastless tracks, and it is difficult to effectively reduce the risk of cracking. In summary, the existing technologies still have problems such as cumbersome operation, strong subjectivity, and low evaluation accuracy in practical applications. There is an urgent need for a more advanced crack status evaluation method based on automated processing to improve track maintenance efficiency. Summary of the invention

[0004] The purpose of the present invention is to provide a method and device for evaluating the crack state of a double-block ballastless track to improve the above-mentioned problem. In order to achieve the above-mentioned purpose, the technical solution adopted by the present invention is as follows:

[0005] In a first aspect, the present application provides a method for evaluating crack status of a dual-block ballastless track, comprising:

[0006] Acquire image information of the dual-block ballastless track in real time, wherein the image information includes image information of both sides and top of the dual-block ballastless track;

[0007] Sending the image information of the dual-block ballastless track to a preset image processing module for target detection to obtain crack image information and crack size information of the dual-block ballastless track;

[0008] Determining crack state evaluation index information based on the crack image information and crack size information of the twin-block ballastless track, wherein the crack state evaluation index information includes a level and a quantity of the crack state evaluation index;

[0009] The crack state evaluation index information is sent to a preset crack state evaluation module for processing, wherein the evaluation index weight calculation and crack state evaluation are performed in sequence through a preset calculation formula to obtain a state evaluation result of the dual-block ballastless track.

[0010] In a second aspect, the present application also provides a dual-block ballastless track crack state evaluation device, comprising:

[0011] An acquisition unit, used for acquiring image information of the dual-block ballastless track in real time, wherein the image information includes image information of both sides and top of the dual-block ballastless track;

[0012] A detection unit, used for sending the image information of the dual-block ballastless track to a preset image processing module for target detection, so as to obtain crack image information and crack size information of the dual-block ballastless track;

[0013] A processing unit, configured to determine crack state evaluation index information based on the crack image information and crack size information of the bi-block ballastless track, wherein the crack state evaluation index information includes a level and a quantity of the crack state evaluation index;

[0014] The evaluation unit is used to send the crack state evaluation index information to a preset crack state evaluation module for processing, wherein the evaluation index weight calculation and crack state evaluation are performed in sequence through a preset calculation formula to obtain the state evaluation result of the dual-block ballastless track.

[0015] The beneficial effects of the present invention are:

[0016] The present invention realizes real-time monitoring and accurate evaluation of the crack status of dual-block ballastless track by introducing advanced technologies such as image processing, target detection and complexity analysis. The method of the present invention obtains the image information of the dual-block ballastless track, classifies it, and constructs a Fibonacci heap for thread management. It combines Fourier transform to obtain the frequency domain characteristics of the image, and then distributes the number of images according to the complexity of the crack image, thereby ensuring the efficiency and accuracy of the crack detection process. At the same time, the present invention obtains the texture image of the crack through grayscale transformation and linear interpolation processing, and performs edge detection on it to accurately calculate the crack size. Based on the multi-level indicator evaluation of crack image information and size information, weighted scoring and membership function are used to finally give a crack status score, which provides a scientific basis for crack monitoring and maintenance of dual-block ballastless track. Compared with the prior art, the present invention significantly improves the accuracy and objectivity of the evaluation results while improving the detection efficiency.

[0017] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or be understood by implementing the embodiments of the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description, claims, and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0019] Figure 1 It is a schematic flow chart of a method for evaluating crack status of a dual-block ballastless track according to an embodiment of the present invention;

[0020] Figure 2 Schematic diagram of the structure of a dual-block ballastless track crack state evaluation device described in an embodiment of the present invention.

[0021] In the figure: 701, acquisition unit; 702, detection unit; 703, processing unit; 704, evaluation unit; 7021, first detection subunit; 7022, second detection subunit; 7023, third detection subunit; 7024, fourth detection subunit; 7025, fifth detection subunit; 70251, first recognition subunit; 70252, second recognition subunit; 70253, third recognition subunit; 70254, fourth recognition subunit; 7031, first processing subunit; 7032, second processing subunit; 7041, first evaluation subunit; 7042, second evaluation subunit; 7043, third evaluation subunit; 70431, first calculation subunit; 70432, second calculation subunit; 70433, third calculation subunit. DETAILED DESCRIPTION

[0022] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0023] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0024] Embodiment 1:

[0025] This embodiment provides a method for evaluating the crack status of a dual-block ballastless track.

[0026] See also Figure 1 , the figure shows that the method includes step S1, step S2, step S3 and step S4.

[0027] Step S1, acquiring image information of the dual-block ballastless track in real time, wherein the image information includes image information of both sides and top of the dual-block ballastless track;

[0028] It can be understood that the real-time acquisition of image information of the dual-block ballastless track in this step is the basis of the entire crack state evaluation process. Here, it is not just a simple image acquisition, but also the multi-angle and multi-dimensional data requirements in the actual application scenario must be considered. In practical applications, due to structural differences and environmental influences, different types of cracks may exist on both sides and the top of the dual-block ballastless track. For example, the detection difficulty of the through cracks on the top and the eight-shaped cracks on the side is different. Therefore, by collecting data from multiple angles through high-resolution cameras or other image acquisition devices, comprehensive monitoring of different crack types can be ensured. In addition, considering the influence of train running speed and environmental factors (such as light and weather), real-time acquisition means that acquisition equipment with high-speed data processing capabilities must be used to ensure that clear and accurate image information can still be obtained under high-speed running conditions of high-speed rail. Therefore, this device has a total of 3 high-definition cameras to perform high-definition imaging of the sleepers, roadbeds, and base surfaces of the dual-block ballastless track, and identify cracks and other diseases. HD cameras 1 and 2 are installed on both sides to image the sides of the ballastless track, and HD camera 3 is installed in the middle to image the surface of the ballastless track bed. Two wheels are installed under the detection equipment, which can be placed on the rails and move back and forth. The wheels are connected to the motor and the motor controller. The inspector can realize mobile detection of the equipment trolley by remotely controlling the motor controller.

[0029] Step S2, sending the image information of the dual-block ballastless track to a preset image processing module for target detection, and obtaining crack image information and crack size information of the dual-block ballastless track;

[0030] It can be understood that this step can quickly and accurately separate the crack information in the original image, automatically generate crack size data, and greatly reduce the subjective error of manual detection. This processing method not only improves the detection efficiency, but also enhances the accuracy of crack assessment, so that track maintenance can be planned and repaired more targeted. In this step, step S2 includes step S21, step S22, step S23, step S24 and step S25.

[0031] Step S21, classifying the image information of the dual-block ballastless track according to the corresponding acquisition time periods, wherein each time period is regarded as a thread, and the occurrence order of each time period is regarded as the thread priority, so as to obtain at least two categories of image information of the dual-block ballastless track;

[0032] It can be understood that this step achieves efficient management and processing of crack image information through the classification of time periods and the setting of thread priorities. This not only ensures the efficiency of parallel processing of data, but also ensures the timely capture of crack development in key time periods, laying the foundation for the subsequent accurate assessment of crack status. Compared with traditional linear processing methods, this design significantly improves the processing speed and enhances the sensitivity to the dynamic evolution of cracks.

[0033] Step S22: construct a Fibonacci heap with image information of all categories of dual-block ballastless tracks, and insert each thread into the heap as a node, wherein each node carries priority and status information of the thread, to obtain a thread management structure;

[0034] It can be understood that this step can significantly improve the scheduling efficiency of multi-threaded tasks by using the Fibonacci heap for thread management, especially when the priority needs to be adjusted frequently. This step also ensures that the image processing tasks in different time periods dynamically adjust the processing order according to actual needs through this structure, making the system more flexible and adaptable to complex environmental changes. Compared with the traditional linear scheduling algorithm, this method is more efficient in complex task scheduling and can significantly improve the overall performance of the image processing system.

[0035] Step S23, performing Fourier transform on the image information of the dual-block ballastless track in each thread, and calculating the complexity of the image information of each ballastless track based on the frequency domain features obtained by the transform;

[0036] It can be understood that this step converts the image from the spatial domain to the frequency domain to extract its frequency features. The main role of Fourier transform in image processing is to decompose the complex image structure into a series of sine and cosine waves of different frequencies. The frequency domain features can reflect the degree of detail change and periodic structure in the image. In the double-block ballastless track crack detection scenario, the details and complex forms of the cracks are manifested as high-frequency components in the frequency domain, while the background noise and other low-frequency information may reflect the overall structure of the track. After Fourier transform, the complexity of each image can be evaluated by analyzing the frequency domain features of the image. Complexity is an indicator of the amount of image information, usually measured by the distribution and intensity of high-frequency components. The contour, distribution and direction of the crack will appear more obvious in the high-frequency area of ​​the frequency domain. The more complex the image is, the higher the proportion of high-frequency components in its spectrum. Therefore, the complexity of the image can be calculated by counting the proportion and intensity changes of the high-frequency components. This process actually converts the spatial features in the image into frequency domain features in mathematical form, and analyzes these frequency domain features to quantify the complexity of the image.

[0037] The calculation formula of complexity is as follows:

[0038]

[0039] Among them, w j is the width of the jth crack, d j is the depth of the jth crack, K is the number of cracks, k is a normalization coefficient, and D is the complexity.

[0040] Step S24: reallocate the number of images of each thread based on the complexity of the image information of each ballastless track crack, and obtain the image information to be processed of each thread;

[0041] It is understandable that threads with high complexity in this step will be allocated fewer images so that the system can have more resources to process these highly complex images. Threads with lower complexity will be allocated more images because the processing difficulty of these images is relatively low and can be completed quickly. By redistributing images based on complexity, the balance and efficiency of image processing are ensured, so that system resources can be maximized. This allocation method enables high-complexity images to be processed by the system in a priority and more refined manner, thereby improving the accuracy and reliability of crack detection. Compared with the traditional average allocation method, this step can be adaptively adjusted according to the complexity of the image information, so that the system can flexibly respond to the processing requirements of different images, especially when processing high-frequency detail cracks, showing better performance.

[0042] Step S25: Based on a preset image feature extraction module, the image information of each thread is identified and processed to obtain the image information and crack size information of the dual-block ballastless track in each thread.

[0043] It can be understood that this step can effectively identify crack features in complex backgrounds and accurately extract crack size information by combining deep learning algorithms with traditional image processing methods. Compared with traditional image processing methods, this method has the advantages of strong adaptability and high feature extraction accuracy, and is particularly suitable for complex and changeable crack scenes of ballastless tracks. The use of image feature extraction modules greatly improves the efficiency and accuracy of crack identification, ensuring the accuracy and reliability of subsequent crack status assessment. In this step, step S25 includes step S251, step S252, step S253 and step S254.

[0044] Step S251: performing image recognition on the image information of each thread, extracting each frame of the image containing the dual-block ballastless track image, and obtaining at least two frames of the dual-block ballastless track image;

[0045] It can be understood that this step performs image recognition on the image information of each thread, with the purpose of extracting images related to the dual-block ballastless track from each frame of the image and ensuring that the extracted images can be used for subsequent crack detection and analysis. The core task of image recognition is to screen out images with track structures while excluding irrelevant image interference to ensure the effectiveness of subsequent processing.

[0046] Step S252: grayscale transformation is performed on all dual-block ballastless track images to obtain a grayscale image corresponding to the dual-block ballastless track image, and pixel points with the same grayscale value in the grayscale image are connected by lines, wherein the lines are interpolated by a linear interpolation method to obtain a crack texture image of the dual-block ballastless track;

[0047] It can be understood that this step can enhance the visualization effect of cracks through the combination of grayscale transformation and linear interpolation, so that the originally complex color or noisy image information can be simplified and refined. The generation of grayscale images greatly reduces the computational complexity of the processing process, and the application of interpolation methods ensures the integrity and accuracy of the crack texture. Under this treatment, the characteristics of the cracks are highlighted, and the system can perform subsequent detection and evaluation of cracks more accurately, effectively improving the accuracy of crack identification and the efficiency of image processing.

[0048] Step S253, performing edge detection on the crack texture image of the dual-block ballastless track to obtain a crack contour edge in the ballastless track image, and converting the image outside the crack contour edge into white to obtain a crack contour edge image of the dual-block ballastless track;

[0049] It can be understood that in this step, the crack contour is accurately extracted through edge detection and background transformation steps, ensuring the accuracy and visualization of the crack shape. At the same time, the whitening of the background further reduces the noise and interference in the image, and improves the efficiency of the system in processing crack information. In practical applications, this step significantly improves the accuracy of crack detection and contour extraction, and ensures the accuracy and consistency of crack size and morphology information in the subsequent evaluation process. This provides reliable basic data for the final crack state evaluation and effectively enhances the robustness of the detection system.

[0050] Step S254: Calculate the size of the crack based on the contour edge image of the dual-block ballastless track to obtain crack size information in the crack contour edge image of the dual-block ballastless track.

[0051] It is understood that the calculation of the crack size in this step requires the analysis of the geometric shape of each crack contour in the edge detection image. Based on image processing, some geometric algorithms are usually used to measure the size of the crack. Since the crack contour has been clearly presented through edge detection, the system can use these contour points for calculation. For example, a common method is to determine the length of the crack by measuring the farthest distance of the crack contour, or to use the minimum circumscribed rectangle of the crack edge to estimate the width of the crack.

[0052] Specifically, the crack length can be calculated by measuring the pixel distance from the crack start point to the end point. This distance is usually accurately calculated using the Euclidean distance, that is, the total length of the crack is accumulated by adding the straight-line distance between adjacent pixel points on the crack contour. The crack width can be determined by calculating the maximum distance between the edges of the crack or the width of the minimum circumscribed rectangle. This method ensures the accuracy of the crack width and is particularly suitable for irregular cracks.

[0053] Step S3, determining crack state evaluation index information based on the crack image information and crack size information of the twin-block ballastless track, wherein the crack state evaluation index information includes the level and quantity of the crack state evaluation index;

[0054] It can be understood that in this step, step S3 includes step S31 and step S32.

[0055] Step S31, segmenting the crack image information and crack size information of the dual-block ballastless track based on a preset image processing range to obtain at least one evaluation unit, wherein each evaluation unit contains all crack image information and crack size information within a preset image processing range, and the preset image processing range is a preset length range of the dual-block ballastless track;

[0056] It can be understood that this step greatly enhances the accuracy and flexibility of the system through segmented processing. By dividing the evaluation unit, the system can perform detailed analysis on the local area of ​​the track, thereby more comprehensively evaluating the crack condition of the entire track. Especially in long-distance tracks, the crack distribution may vary greatly. Segmented processing can better capture such local differences and prevent potential risks in certain key areas from being ignored due to averaging. At the same time, this segmented processing method makes large-scale data processing more efficient, reduces the complexity of overall data calculation, and helps to improve the response speed and accuracy of the real-time crack detection system.

[0057] Step S32: performing index conversion on the crack image information and crack size information in all evaluation units based on the preset evaluation index categories, and determining the index levels and index numbers of all evaluation units based on the converted indexes.

[0058] It can be understood that this step simplifies a large amount of complex crack information into intuitive evaluation criteria through reasonable indicator conversion and classification processing. This evaluation method based on quantitative analysis can significantly improve the efficiency of crack detection and evaluation, and due to the adjustability of its evaluation criteria, it can also be adapted and optimized according to specific track types and environmental conditions. This ensures that the system can obtain accurate and reliable evaluation results under different circumstances, thereby providing reliable safety guarantees for the daily operation of the railway. The conversion method is through the state evaluation system, in which the first-level indicator is the crack state of the CRTS double-block ballastless track, the second-level indicators are sleepers, roadbed plates and base plates, and the third-level indicators are indicators such as sleeper cracks, roadbed plate cracks, sleeper interface cracks and base plate cracks.

[0059] Step S4: Send the crack state evaluation index information to a preset crack state evaluation module for processing, wherein the evaluation index weight calculation and crack state evaluation are performed in sequence by a preset calculation formula to obtain a state evaluation result of the dual-block ballastless track.

[0060] It can be understood that this step uses a preset crack state evaluation module to comprehensively process the extracted and converted crack state evaluation index information, and finally generates the state evaluation result of the track. In this step, the weight calculation of the evaluation index is the key, which not only affects the accuracy of the final evaluation, but also determines the importance of each crack feature in the overall evaluation. In this process, the system first assigns weights to the indicators of each evaluation unit according to the preset calculation formula. Usually, this weight is determined by analyzing the impact of each crack feature (such as crack length, width, depth, etc.) on the overall safety of the track. In this step, step S4 includes step S41, step S42 and step S43.

[0061] Step S41, scoring the indicators of the same level in the crack state evaluation indicator information, and constructing a judgment matrix based on the obtained scores;

[0062] It can be understood that in this step, first of all, for each indicator of the same level in the crack status evaluation index information, it is necessary to score according to its importance to track safety. This score is usually collected by multiple experts in the field through collective discussion or questionnaires, taking into account the impact of different crack characteristics (such as crack length, width, depth and its position in the track structure) on the overall track health. The expert scoring process is usually iterative, and a relatively consistent score is finally obtained through a comprehensive evaluation of various indicators. This method can not only introduce the rich experience and expertise of experts, but also reduce subjective bias to a certain extent and improve the reliability of scoring.

[0063] After collecting the scores of indicators at the same level, the system will construct a judgment matrix based on these scores. The judgment matrix is ​​a tool in decision analysis, which is in the form of a square matrix. Each element in the matrix represents the relative importance of a pair of indicators. Specifically, if a certain indicator is more important than another indicator in the safety assessment, its corresponding matrix element will give a value greater than 1, otherwise it will be less than 1. In order to ensure the rationality of the judgment matrix, it is usually necessary to meet the consistency principle, that is, in the comparison process, if A is more important than B, and B is more important than C, then A should be more important than C. The implementation of this principle can be evaluated by the consistency ratio (CR) to ensure the logical consistency of the judgment matrix.

[0064] The judgment matrix constructed in this way not only provides a reliable basis for the subsequent weight calculation, but also reflects the relative importance of each indicator in the evaluation, which in turn affects the final evaluation result of the entire crack state. This systematic scoring and judgment matrix construction method improves the scientificity and repeatability of crack state evaluation, ensuring that the final output evaluation results are more authoritative and practical. The ultimate technical effect is to provide a clear and reasonable framework for the crack state evaluation of dual-block ballastless track, enhance the transparency and accuracy of the decision-making process, and provide strong support for track maintenance and safety management.

[0065] Step S42, calculating the weight vector of the judgment matrix based on the arithmetic mean method to obtain the weight vector of the judgment matrix of each level of indicators;

[0066] It can be understood that this step can more clearly identify the most critical indicators for track safety, so that these key indicators can be given priority in subsequent status assessment and decision-making. This not only improves the efficiency of the assessment, but also provides a more accurate reference for track maintenance management, which helps to discover potential risks in advance and take corresponding measures to ensure the safe operation of the track. Among them, the formula for calculating the weight vector of the judgment matrix by the arithmetic mean method is as follows:

[0067]

[0068] Where n is the total number of indicators, a ij is the relative importance of indicator i to indicator j in the judgment matrix, w i The relative importance of each indicator in crack state assessment.

[0069] Step S43: Calculate the membership of the crack state evaluation index information and the preset evaluation score, and determine the final crack state score of the twin-block ballastless track based on the evaluation score of each level of the index calculated.

[0070] It can be understood that this step makes the crack status score more scientific and comprehensive through the method of fuzzy logic and weight integration. Compared with the traditional method, which only relies on a single indicator or subjective evaluation, the membership calculation method can integrate multiple aspects of information and overcome the deviation caused by subjective factors, thereby improving the accuracy and reliability of the score. In addition, the final crack status score provides a quantitative basis for the maintenance and management of the track, which helps to formulate a more reasonable inspection and maintenance strategy to ensure the safety and stability of railway operations. In this step, step S43 includes step S431, step S432 and step S433.

[0071] Step S431, constructing a membership function based on all the crack state evaluation index information, and calculating the membership value corresponding to each crack state evaluation index information;

[0072] It can be understood that this step can transform the original quantitative crack state into a qualitative fuzzy evaluation by constructing a membership function and calculating the membership value. The advantage of this method is that it can take into account the uncertainty and fuzziness of the indicators, making the evaluation results more objective and scientific. Compared with the traditional method that only relies on the judgment of absolute values, fuzzy evaluation provides a more flexible way to make reasonable evaluations in complex situations. The introduction of this fuzzy logic not only improves the accuracy of crack state evaluation, but also provides strong data support for subsequent decision-making.

[0073] The membership function is as follows:

[0074]

[0075] Among them, S, S1, S2, and S3 represent the upper thresholds corresponding to the indicators of the four levels, f(s) represents the membership value, and d represents c means

[0076] Step S432: Multiply the weight vector of the indicator of each level and its corresponding membership value to obtain the evaluation score of the indicator of each level;

[0077] It can be understood that this step realizes the quantitative analysis of crack status evaluation by combining weights and membership. This processing method has multiple advantages: first, it allows the importance of different indicators to be flexibly adjusted to better reflect the actual situation; second, by combining fuzzy evaluation with quantitative calculation, it can effectively reduce the risk of misjudgment caused by a single indicator and improve the reliability and accuracy of the overall evaluation. In addition, the final evaluation score is not only easy to understand and analyze, but also provides a scientific basis for decision makers when maintaining and managing double-block ballastless track, ensuring more effective resource allocation and risk control.

[0078] Step S433: construct an evaluation matrix based on the evaluation scores of the indicators of each level, and select the smallest score in the matrix as the final evaluation result, which is the crack state score of the twin-block ballastless track.

[0079] It can be understood that this step can effectively integrate the evaluation results of various indicators by constructing an evaluation matrix and selecting the minimum score, reflecting the weakest link of the dual-block ballastless track in a cracked state. The technical effect of this method is that it ensures that the evaluation results are not affected by the extreme values ​​of a particular indicator, thereby improving the robustness and reliability of the results. At the same time, by emphasizing the selection of the minimum score, the potential risk points in the crack state assessment are highlighted, providing an important basis for subsequent maintenance and management decisions. This systematic approach not only improves the scientific nature of crack state assessment, but also provides decision makers with a clear direction for improvement, ensuring the safety and operational efficiency of the track.

[0080] Embodiment 2:

[0081] like Figure 2 As shown, this embodiment provides a double-block ballastless track crack state evaluation device, see Figure 2 The device includes an acquisition unit 701 , a detection unit 702 , a processing unit 703 and an evaluation unit 704 .

[0082] An acquisition unit 701 is used to acquire image information of the dual-block ballastless track in real time, wherein the image information includes image information of both sides and top of the dual-block ballastless track;

[0083] A detection unit 702 is used to send the image information of the dual-block ballastless track to a preset image processing module for target detection, so as to obtain crack image information and crack size information of the dual-block ballastless track;

[0084] The detection unit 702 includes a first detection subunit 7021 , a second detection subunit 7022 , a third detection subunit 7023 , a fourth detection subunit 7024 and a fifth detection subunit 7025 .

[0085] The first detection subunit 7021 is used to classify the image information of the dual-block ballastless track according to its corresponding acquisition time period, wherein each time period is regarded as a thread, and the occurrence order of each time period is regarded as the thread priority, so as to obtain at least two categories of image information of the dual-block ballastless track;

[0086] The second detection subunit 7022 is used to construct a Fibonacci heap with the image information of all types of dual-block ballastless tracks, and insert each thread into the heap as a node, each node carrying the priority and state information of the thread, to obtain a thread management structure;

[0087] The third detection subunit 7023 is used to perform Fourier transform based on the image information of the dual-block ballastless track in each thread, and calculate the complexity of the image information of each ballastless track based on the frequency domain features obtained by the transform;

[0088] The fourth detection subunit 7024 is used to reallocate the number of images of each thread based on the complexity of the image information of each ballastless track crack, and obtain the image information to be processed of each thread;

[0089] The fifth detection subunit 7025 is used to identify and process the image information of each thread based on a preset image feature extraction module to obtain the image information and crack size information of the dual-block ballastless track in each thread.

[0090] The fifth detection subunit 7025 includes a first recognition subunit 70251 , a second recognition subunit 70252 , a third recognition subunit 70253 and a fourth recognition subunit 70254 .

[0091] The first recognition subunit 70251 is used to perform image recognition on the image information of each thread, extract each frame of the image containing the dual-block ballastless track image, and obtain at least two frames of the dual-block ballastless track image;

[0092] The second identification subunit 70252 is used to perform grayscale transformation on all the dual-block ballastless track images to obtain a grayscale image corresponding to the dual-block ballastless track image, and connect pixel points with the same grayscale value in the grayscale image, wherein the connection is interpolated by a linear interpolation method to obtain a crack texture image of the dual-block ballastless track;

[0093] The third identification subunit 70253 is used to perform edge detection on the crack texture image of the dual-block ballastless track to obtain the crack contour edge in the ballastless track image, and convert the image outside the crack contour edge into white to obtain the crack contour edge image of the dual-block ballastless track;

[0094] The fourth identification subunit 70254 is used to perform size calculation based on the contour edge image of the dual-block ballastless track to obtain crack size information in the crack contour edge image of the dual-block ballastless track.

[0095] A processing unit 703 is used to determine crack state evaluation index information based on the crack image information and crack size information of the dual-block ballastless track, wherein the crack state evaluation index information includes a level and a quantity of the crack state evaluation index;

[0096] The processing unit 703 includes a first processing sub-unit 7031 and a second processing sub-unit 7032 .

[0097] The first processing subunit 7031 is used to segment the crack image information and crack size information of the dual-block ballastless track based on a preset image processing range to obtain at least one evaluation unit, wherein each evaluation unit contains all crack image information and crack size information within a preset image processing range, and the preset image processing range is a preset length range of the dual-block ballastless track;

[0098] The second processing subunit 7032 is used to perform index conversion on the crack image information and crack size information in all evaluation units based on the preset evaluation index category, and determine the index level and index number of all evaluation units based on the converted index.

[0099] The evaluation unit 704 is used to send the crack state evaluation index information to a preset crack state evaluation module for processing, wherein the evaluation index weight calculation and crack state evaluation are performed in sequence through a preset calculation formula to obtain the state evaluation result of the dual-block ballastless track.

[0100] The evaluation unit 704 includes a first evaluation subunit 7041 , a second evaluation subunit 7042 and a third evaluation subunit 7043 .

[0101] The first evaluation subunit 7041 is used to score the indicators of the same level in the crack state evaluation indicator information, and construct a judgment matrix based on the obtained scores;

[0102] The second evaluation subunit 7042 is used to calculate the weight vector of the judgment matrix based on the arithmetic mean method to obtain the weight vector of the judgment matrix of each level of indicators;

[0103] The third evaluation subunit 7043 is used to calculate the membership of the crack state evaluation index information and the preset evaluation score, and determine the final crack state score of the twin-block ballastless track based on the evaluation score of each level of the index calculated.

[0104] The third evaluation subunit 7043 includes a first calculation subunit 70431 , a second calculation subunit 70432 and a third calculation subunit 70433 .

[0105] The first calculation subunit 70431 is used to construct a membership function based on all the crack state evaluation index information, and calculate the membership value corresponding to each crack state evaluation index information;

[0106] The second calculation subunit 70432 is used to perform multiplication calculation based on the weight vector of the indicator of each level and its corresponding membership value to obtain the evaluation score of the indicator of each level;

[0107] The third calculation subunit 70433 is used to construct an evaluation matrix based on the evaluation scores of the indicators of each level, and select the smallest score in the matrix as the final evaluation result, which is the crack state score of the twin-block ballastless track.

[0108] It should be noted that, regarding the device in the above embodiment, the specific manner in which each module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.

[0109] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

[0110] The above is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A method for evaluating crack status of a dual-block ballastless track, characterized in that: include: Acquire image information of the dual-block ballastless track in real time, wherein the image information includes image information of both sides and top of the dual-block ballastless track; Sending the image information of the dual-block ballastless track to a preset image processing module for target detection to obtain crack image information and crack size information of the dual-block ballastless track; Determining crack state evaluation index information based on the crack image information and crack size information of the twin-block ballastless track, wherein the crack state evaluation index information includes a grade and a quantity of the crack state evaluation index; The crack state evaluation index information is sent to a preset crack state evaluation module for processing, wherein the evaluation index weight calculation and crack state evaluation are performed in sequence through a preset calculation formula to obtain a state evaluation result of the dual-block ballastless track.

2. The method for evaluating crack status of a twin-block ballastless track according to claim 1, characterized in that , sending the image information of the dual-block ballastless track to a preset image processing module for target detection, including: Classifying the image information of the dual-block ballastless track according to the corresponding acquisition time periods, wherein each time period is regarded as a thread, and the occurrence order of each time period is regarded as the thread priority, to obtain at least two categories of image information of the dual-block ballastless track; A Fibonacci heap is constructed with the image information of all categories of double-block ballastless tracks, and each thread is inserted into the heap as a node. Each node carries the priority and state information of the thread, and a thread management structure is obtained. Performing Fourier transformation on the image information of the dual-block ballastless track in each thread, and calculating the complexity of the image information of each ballastless track based on the frequency domain features obtained by the transformation; The number of images of each thread is redistributed based on the complexity of the image information of each ballastless track crack, and the image information to be processed by each thread is obtained; Based on a preset image feature extraction module, the image information of each thread is identified and processed to obtain the crack image information and crack size information of the dual-block ballastless track in each thread.

3. The method for evaluating crack status of a twin-block ballastless track according to claim 2, characterized in that ,Based on the preset image feature extraction module, the image information of each thread is recognized and processed, including: Performing image recognition on the image information of each thread, extracting each frame of the image containing the dual-block ballastless track image, and obtaining at least two frames of the dual-block ballastless track image; Performing grayscale transformation on all dual-block ballastless track images to obtain grayscale images corresponding to the dual-block ballastless track images, and connecting pixel points with the same grayscale value in the grayscale images, wherein a linear interpolation method is used to interpolate the connecting lines to obtain a crack texture image of the dual-block ballastless track; Performing edge detection on the crack texture image of the dual-block ballastless track to obtain a crack contour edge in the ballastless track image, and converting the image outside the crack contour edge into white to obtain a crack contour edge image of the dual-block ballastless track; The size calculation is performed based on the contour edge image of the dual-block ballastless track to obtain the crack size information in the crack contour edge image of the dual-block ballastless track.

4. The method for evaluating crack status of a twin-block ballastless track according to claim 1, characterized in that , based on the crack image information and crack size information of the dual-block ballastless track, determining crack state evaluation index information, including: Segmenting the crack image information and crack size information of the dual-block ballastless track based on a preset image processing range to obtain at least one evaluation unit, wherein each evaluation unit contains all crack image information and crack size information within a preset image processing range, and the preset image processing range is a preset length range of the dual-block ballastless track; Based on the preset evaluation index categories, the crack image information and crack size information in all evaluation units are converted into indicators, and the indicator levels and the number of indicators of all evaluation units are determined based on the converted indicators.

5. The method for evaluating crack status of a twin-block ballastless track according to claim 1, characterized in that ,The crack state evaluation index information is sent to the preset crack state evaluation module for processing, including: Scoring the indicators of the same level in the crack state evaluation indicator information, and constructing a judgment matrix based on the obtained scores; The weight vector of the judgment matrix is ​​calculated based on the arithmetic mean method to obtain the weight vector of the judgment matrix of each level of indicators; The crack state evaluation index information and the preset evaluation score are calculated for membership, and the final crack state score of the twin-block ballastless track is determined based on the evaluation score of each level of the index calculated.

6. A device for evaluating crack status of a double-block ballastless track, characterized in that: include: An acquisition unit, used for acquiring image information of the dual-block ballastless track in real time, wherein the image information includes image information of both sides and top of the dual-block ballastless track; A detection unit, used for sending the image information of the dual-block ballastless track to a preset image processing module for target detection, so as to obtain crack image information and crack size information of the dual-block ballastless track; A processing unit, configured to determine crack state evaluation index information based on the crack image information and crack size information of the bi-block ballastless track, wherein the crack state evaluation index information includes a level and a quantity of the crack state evaluation index; The evaluation unit is used to send the crack state evaluation index information to a preset crack state evaluation module for processing, wherein the evaluation index weight calculation and crack state evaluation are performed in sequence through a preset calculation formula to obtain the state evaluation result of the dual-block ballastless track.

7. The device for evaluating crack status of a twin-block ballastless track according to claim 6, characterized in that: The detection unit comprises: A first detection subunit is used to classify the image information of the dual-block ballastless track according to its corresponding acquisition time period, wherein each time period is regarded as a thread, and the occurrence order of each time period is regarded as the thread priority, so as to obtain at least two categories of image information of the dual-block ballastless track; The second detection subunit is used to construct a Fibonacci heap with image information of all categories of dual-block ballastless tracks, and insert each thread into the heap as a node, each node carrying priority and state information of the thread, to obtain a thread management structure; A third detection subunit is used to perform Fourier transform based on the image information of the dual-block ballastless track in each thread, and calculate the complexity of the image information of each ballastless track based on the frequency domain features obtained by the transform; The fourth detection subunit is used to reallocate the number of images of each thread based on the complexity of the image information of each ballastless track crack, and obtain the image information to be processed of each thread; The fifth detection subunit is used to identify and process the image information of each thread based on a preset image feature extraction module to obtain the crack image information and crack size information of the dual-block ballastless track in each thread.

8. The device for evaluating crack status of a twin-block ballastless track according to claim 7, characterized in that: The fifth detection subunit comprises: The first recognition subunit is used to perform image recognition on the image information of each thread, extract each frame of the image containing the dual-block ballastless track image, and obtain at least two frames of the dual-block ballastless track image; The second identification subunit is used to perform grayscale transformation on all the dual-block ballastless track images to obtain a grayscale image corresponding to the dual-block ballastless track image, and connect pixel points with the same grayscale value in the grayscale image, wherein the connection is interpolated by a linear interpolation method to obtain a crack texture image of the dual-block ballastless track; A third recognition subunit is used to perform edge detection on the crack texture image of the dual-block ballastless track to obtain a crack contour edge in the ballastless track image, and convert the image outside the crack contour edge into white to obtain a crack contour edge image of the dual-block ballastless track; The fourth identification subunit is used to perform size calculation based on the contour edge image of the dual-block ballastless track to obtain crack size information in the crack contour edge image of the dual-block ballastless track.

9. The device for evaluating crack status of a twin-block ballastless track according to claim 6, characterized in that: The processing unit comprises: a first processing subunit, configured to segment the crack image information and crack size information of the dual-block ballastless track based on a preset image processing range to obtain at least one evaluation unit, wherein each evaluation unit contains all crack image information and crack size information within a preset image processing range, and the preset image processing range is a preset length range of the dual-block ballastless track; The second processing subunit is used to perform index conversion on the crack image information and crack size information in all evaluation units based on preset evaluation index categories, and determine the index levels and index numbers of all evaluation units based on the converted indexes.

10. The device for evaluating crack status of a twin-block ballastless track according to claim 6, characterized in that: The evaluation unit comprises: A first evaluation subunit is used to score the indicators of the same level in the crack state evaluation indicator information, and to construct a judgment matrix based on the obtained scores; The second evaluation subunit is used to calculate the weight vector of the judgment matrix based on the arithmetic mean method to obtain the weight vector of the judgment matrix of each level of indicators; The third evaluation subunit is used to calculate the membership of the crack state evaluation index information and the preset evaluation score, and determine the final crack state score of the dual-block ballastless track based on the evaluation score of each level of the index calculated.

Images