Pavement condition rapid detection method based on two-way stereoscopic vision

Abstract

The invention discloses a road surface condition rapid detection method based on two-way stereoscopic vision, and relates to the field of rapid detection, and the method comprises the following modules: a three-dimensional model optimization and texture mapping module, which is used for carrying out smoothing processing and detail enhancement operation on a three-dimensional model by adopting a model optimization algorithm, and meanwhile, carrying out texture mapping on the three-dimensional model; according to the method, the three-dimensional model optimization and texture mapping module adopts a model optimization algorithm and a texture mapping technology, the visual effect and authenticity of the three-dimensional model are improved, the model is enabled to be more vivid by performing smoothing processing and detail enhancement operation on the model, and the image quality is improved. Meanwhile, the image information of the road surface is attached to the surface of the model through the texture mapping technology, the visual expressive force of the model is enhanced, the visual quality of the three-dimensional model is improved, and a user can know the actual condition of the road surface more visually.

Description

Technical Field

[0001] The present invention relates to the field of rapid detection, and in particular to a method for rapid detection of road conditions based on dual-path stereoscopic vision. Background Art

[0002] With the vigorous development of intelligent transportation systems, accurate and real-time detection of road conditions has become a key link in ensuring traffic safety and improving road maintenance efficiency. The traditional road detection method is difficult to ensure the objectivity and accuracy of the detection results; it is often unable to fully and accurately reflect the real condition of the road. In recent years, the rapid development of computer vision technology has provided new solutions for road detection. The use of dual-channel stereo vision technology can obtain the three-dimensional geometric information and depth information of objects by simulating the binocular vision of the human eye, thereby providing more comprehensive and accurate data support for road condition detection.

[0003] However, when the existing road surface detection methods based on dual-path stereo vision detect complex and changeable road conditions, factors such as lighting conditions, texture characteristics or occlusion will affect the performance of the dual-path stereo vision system, resulting in the accuracy and stability of the detection results being affected. Secondly, most of the existing detection methods use complex algorithms and models, which have large computational complexity and poor real-time performance, and are difficult to meet the needs of practical applications. In addition, for the automatic identification and classification of road surface defects, the existing detection methods also lack efficient and accurate methods.

[0004] Therefore, how to construct an efficient road condition rapid detection method based on dual-path stereo vision and a road condition rapid detection method of three-dimensional reconstruction. However, most of the road condition rapid detection methods based on dual-path stereo vision on the market can only meet some of the requirements of the above characteristics, and cannot realize the use of dual-path stereo vision technology to quickly and accurately obtain three-dimensional information of the road surface. Therefore, developing a road condition rapid detection method based on dual-path stereo vision that can comprehensively realize the above characteristics is of great significance for the accurate detection and intuitive display of road conditions, and for providing efficient and accurate road information support for intelligent transportation systems. Summary of the invention

[0005] The purpose of the present invention is to make up for the shortcomings of the existing technology and provide a method for rapid detection of road conditions based on dual-path stereo vision. It can quickly obtain three-dimensional information of the road surface through dual-path stereo vision technology, realize efficient and accurate detection of road conditions, and display the actual situation of the road surface through a variety of visualization methods, provide real-time and accurate road information support for intelligent transportation systems, and improve road maintenance levels and traffic safety.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: a method for rapid detection of road conditions, characterized in that the method comprises the following modules:

[0007] Dual-channel stereo vision data acquisition and processing module: Use the dual-channel stereo vision system to synchronously collect image data of the road surface, extract feature points in the image through image processing technology, and use the stereo matching algorithm to achieve accurate matching of feature points, providing data support for subsequent 3D reconstruction and visualization;

[0008] 3D information extraction and model reconstruction module: Based on the 3D information obtained by the dual-channel stereo vision system, a high-precision 3D model of the road surface is generated using a 3D reconstruction algorithm. Advanced algorithms and technologies are used to ensure the authenticity and accuracy of the model, providing a reliable basis for subsequent road condition analysis.

[0009] 3D model optimization and texture mapping module: The 3D model is smoothed and enhanced in detail by using a model optimization algorithm. At the same time, the model is texture mapped using image texture information to make the model more realistic, thereby improving the visual effect and authenticity of the 3D model and enabling users to more intuitively understand the actual conditions of the road surface.

[0010] Pavement condition visualization and analysis module: The present invention provides a variety of visualization methods including three-dimensional rendering and texture mapping, so that users can intuitively view the three-dimensional model of the road surface. At the same time, the module has an automatic analysis function, which can detect cracks and potholes in the road surface according to the three-dimensional model and generate corresponding inspection reports. Users can view, analyze and export the inspection reports through the interactive interface, so as to timely process and maintain the road condition.

[0011] Furthermore, the dual-channel stereo vision data acquisition and processing module uses a dual-channel camera to synchronously capture images of the road surface from two different angles to obtain two left and right stereo images, ensure that the acquired images contain sufficient three-dimensional scene information, calibrate the dual-channel camera, and obtain internal and external parameters of the camera, such as focal length, principal point coordinates and distortion coefficient, so that in subsequent steps, pixel points in the image can be mapped to points in three-dimensional space, and pre-process the acquired stereo images, including denoising, filtering and edge detection operations to improve image quality and reduce errors in subsequent processing, extract feature points or feature lines from the pre-processed images, use a stereo matching algorithm to match the feature points or lines in the left and right images, and find corresponding points or lines in the left and right images by calculating the similarity or distance between the feature points, thereby establishing a stereo correspondence.

[0012] Furthermore, the dual-channel stereo vision data acquisition and processing module determines the disparity value of each pixel by using the matching cost calculated by the stereo matching algorithm unit, and generates a disparity map, thereby further extracting the three-dimensional information of the road surface. The calculation formula is: Cx,y,d=∑ i∈Np I L x+i,yIR x+i+d,y, where C(x, y, d) represents the matching cost of the pixel point (x, y) when the disparity is d, N(p) is the neighborhood set centered on the pixel point p(x, y), and I L x+i,y and I R x+i+d,y are the grayscale values ​​of the corresponding pixels in the left and right views respectively. Through the stereo matching algorithm, this module can accurately and efficiently process dual-channel stereo vision data to achieve high-precision three-dimensional reconstruction and visualization of the road surface.

[0013] Furthermore, the three-dimensional information extraction and model reconstruction module calculates the disparity of each pixel point based on the result of stereo matching, combines the internal and external parameters of the camera and the disparity information, and uses the triangulation principle to calculate the coordinates of each pixel point in the three-dimensional space. Based on the calculated three-dimensional coordinate points, the three-dimensional model of the road surface is reconstructed using point cloud data or a gridding method, and a smoothing algorithm is used in the reconstruction process to improve the accuracy and smoothness of the model.

[0014] Furthermore, the three-dimensional information extraction and model reconstruction module uses a smoothing algorithm to process the acquired three-dimensional coordinate data to reduce noise and improve model accuracy, wherein the smoothing algorithm uses a moving average method, and its algorithm formula is: Where W is the weight, N=2m+1 is the number of moving average items, X(i) is the variable in window A, and each time window A moves forward by one data point, the variable index j also moves backward by one data point. The three-dimensional coordinate data processed by the smoothing algorithm is used in the subsequent model reconstruction steps to generate a more accurate and smoother three-dimensional model.

[0015] Furthermore, the 3D model optimization and texture mapping module optimizes the reconstructed 3D model, including smoothing filtering and detail enhancement operations, to improve the visual effect and authenticity of the model, uses the image information of the road surface to perform texture mapping on the 3D model, and by selecting appropriate texture images and mapping algorithms, fits the texture information to the surface of the model, making the model more realistic and vivid.

[0016] Furthermore, the three-dimensional model optimization and texture mapping module adopts smoothing filtering and mapping algorithm to improve the visual effect and authenticity of the model, and its smoothing filtering formula is: Among them, h(x, y) is the result after filtering, (i, j) is the coordinate of the signal value to be filtered in the image, σ is the standard deviation of the Gaussian function, which determines the smoothness of the filtering. Texture mapping usually involves coordinate transformation, mapping the coordinates (u, v) of the texture space to the vertex coordinates of the three-dimensional model surface. Through the texture mapping algorithm, the noise and irregularities on the model surface can be effectively removed, and the smoothness and visual effect of the model can be improved. The core formula of the texture mapping algorithm is: Tp=LerpTv1,Tv2,α, among which T(p) is the texture coordinate corresponding to the point p on the model surface, Tv1 and Tv2 are the texture coordinates of the two vertices v1 and v2 adjacent to the point p, and α is the interpolation parameter, which indicates the position of the point p relative to v1 and v2. Through this mapping algorithm, the texture can be smoothly fitted to the model surface, improving the realism and visual effect of the model. At the same time, combined with the three-dimensional model optimization processing, a high-quality and detailed three-dimensional model can be generated, providing strong support for subsequent visualization and analysis.

[0017] Furthermore, the pavement condition visualization and analysis module displays the optimized and texture-mapped three-dimensional model to the user in a variety of visualization methods, such as three-dimensional rendering or texture mapping. The user can zoom, rotate and roam through the interactive interface to fully and intuitively understand the actual situation of the pavement. The three-dimensional model is used for automatic analysis to detect cracks or potholes in the pavement. By calculating the geometric features and color change information of the model, the location, size and severity of the disease can be identified, and a corresponding detection report can be generated.

[0018] Furthermore, the method also includes a calibration module for periodically calibrating the camera to ensure the accuracy and stability of the collected image data.

[0019] Furthermore, the specific steps of the method are:

[0020] (1) System preparation and initialization: Start the 3D reconstruction visualization system based on dual-channel stereo vision, ensure that the dual-channel stereo vision system operates normally, and that the camera is in a suitable position and angle to capture image data of the road surface.

[0021] (2) Dual-channel stereo vision data acquisition and processing: The dual-channel stereo vision system synchronously acquires image data of the road surface to ensure that the images captured by the left and right cameras are consistent in time for subsequent stereo matching. The acquired image data is pre-processed, including denoising and contrast enhancement, to improve image quality and prepare for subsequent feature point extraction and matching.

[0022] (3) Feature point extraction and stereo matching: In the preprocessed image, feature points are extracted using image processing technology. These feature points are usually prominent areas in the image, such as edges and corners. The feature points in the images captured by the left and right cameras are accurately matched using a stereo matching algorithm to obtain the correspondence between the feature points.

[0023] (4) 3D information extraction and model reconstruction: Based on the matching results of feature points, the 3D coordinates of the feature points are calculated through a 3D reconstruction algorithm to obtain the 3D information of the road surface. The 3D information obtained is then used to generate a high-precision 3D model of the road surface through a 3D reconstruction algorithm.

[0024] (5) 3D model optimization and texture mapping: Smooth the generated 3D model to remove noise and irregularities, improve the smoothness and realism of the model, enhance the model’s details to highlight the model’s edges and texture information, and make the model more vivid and realistic. Use the texture information of the road surface image to perform texture mapping on the 3D model, so that the model surface presents richer details and realism.

[0025] (6) Pavement condition visualization and analysis: The optimized 3D model is displayed in the form of 3D rendering, allowing users to intuitively view the 3D structure of the pavement. The system automatically analyzes the 3D model, detects cracks and potholes in the pavement, and generates corresponding inspection reports. Users can view, analyze and export the inspection reports through the interactive interface, and promptly handle and maintain the pavement condition based on the report content.

[0026] Compared with the existing technology, this method for rapid detection of road conditions based on dual-path stereo vision has the following advantages:

[0027] Beneficial effects:

[0028] 1. The present invention adopts model optimization algorithm and texture mapping technology through the three-dimensional model optimization and texture mapping module to improve the visual effect and authenticity of the three-dimensional model, and makes the model more realistic by smoothing and detail enhancement operations on the model; at the same time, the image information of the road surface is attached to the model surface by using texture mapping technology, which enhances the visual expression of the model and improves the visual quality of the three-dimensional model, so that users can understand the actual situation of the road surface more intuitively.

[0029] 2. The present invention adopts an optimized 3D reconstruction algorithm through a 3D information extraction and model reconstruction module to achieve rapid reconstruction of a high-precision 3D model of the road surface, improve the authenticity and detail expression of the model, enable the model to more accurately reflect the actual conditions of the road surface, improve the efficiency and accuracy of 3D reconstruction, and provide a reliable model basis for the analysis and visualization of road conditions.

[0030] Other advantages, objectives and features of the present invention will be set forth in part in the following description and, in part, will be apparent to those skilled in the art based on an examination of the following or may be taught from the practice of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention, and for ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0032] Figure 1 This is a flowchart of a method for rapid detection of road conditions based on dual-path stereo vision. DETAILED DESCRIPTION

[0033] The technical solutions in the embodiments of the present invention are described clearly and completely below. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] Example 1: Stereo matching and 3D information extraction

[0035] A rapid road condition detection method based on dual-path stereo vision, which includes the following modules:

[0036] Dual-channel stereo vision data acquisition and processing module: Use the dual-channel stereo vision system to synchronously collect image data of the road surface, extract feature points in the image through image processing technology, and use the stereo matching algorithm to achieve accurate matching of feature points, providing data support for subsequent 3D reconstruction and visualization;

[0037] 3D information extraction and model reconstruction module: Based on the 3D information obtained by the dual-channel stereo vision system, a high-precision 3D model of the road surface is generated using a 3D reconstruction algorithm. Advanced algorithms and technologies are used to ensure the authenticity and accuracy of the model, providing a reliable basis for subsequent road condition analysis.

[0038] 3D model optimization and texture mapping module: The 3D model is smoothed and enhanced in detail by using a model optimization algorithm. At the same time, the model is texture mapped using image texture information to make the model more realistic, thereby improving the visual effect and authenticity of the 3D model and enabling users to more intuitively understand the actual conditions of the road surface.

[0039] Pavement condition visualization and analysis module: The present invention provides a variety of visualization methods including three-dimensional rendering and texture mapping, so that users can intuitively view the three-dimensional model of the road surface. At the same time, the module has an automatic analysis function, which can detect cracks and potholes in the road surface according to the three-dimensional model and generate corresponding inspection reports. Users can view, analyze and export the inspection reports through the interactive interface, so as to timely process and maintain the road condition.

[0040] In this embodiment, it is described in detail how to use the stereo matching algorithm to extract three-dimensional information from the two-way stereo vision data and reconstruct the three-dimensional model of the road surface. The stereo matching algorithm is used to match the feature points in the left and right images. In this embodiment, the disparity value of each pixel is determined by the matching cost calculated by the stereo matching algorithm unit, and a disparity map is generated to further extract the three-dimensional information of the road surface. The calculation formula is:

[0041]

[0042] Among them, C(x, y, d) represents the matching cost of the pixel point (x, y) when the disparity is d, N(p) is the neighborhood set centered on the pixel point p(x, y), and I L x+i,y and I R x+i+d, y are the grayscale values ​​of the corresponding pixels in the left and right views, respectively. After completing the stereo matching, we obtain the disparity value of each pixel. The disparity value represents the horizontal displacement between the corresponding points in the left and right images, and contains the three-dimensional information of the road surface. According to the internal and external parameters of the camera (such as focal length, principal point coordinates, distortion coefficient) and disparity information, the coordinates of each pixel in the three-dimensional space are calculated using the triangulation principle. The specific calculation formula is:

[0043]

[0044] Among them, Z is the depth information, indicating the distance from the object to the camera, f is the focal length of the camera, which determines the focusing ability of the camera lens, T is the baseline distance between the two cameras, that is, the distance between the optical centers of the two cameras in the stereo vision system, and d is the parallax value, indicating the horizontal displacement between the corresponding points in the left and right camera images. With these three-dimensional coordinate points, the three-dimensional model of the road surface can be reconstructed, and the calculated three-dimensional coordinate points are converted into point cloud data format. The point cloud data is smoothed to remove noise and improve data quality. The gridding method is used to convert the point cloud data into a three-dimensional grid model. This grid model is the three-dimensional reconstruction result of the road surface, which can intuitively display the geometric shape and structure of the road surface. In order to verify the accuracy and reliability of the three-dimensional reconstruction, it can be verified and calibrated. For example, actual measurement data or known road structure information can be used to compare and verify the reconstruction results. In addition, it is also very important to calibrate the camera regularly to ensure the accuracy and stability of the collected image data.

[0045] In summary, through the stereo matching and three-dimensional information extraction process described in this embodiment, two-way stereo vision data can be effectively used to reconstruct the three-dimensional model of the road surface, providing strong support for subsequent road condition analysis, visualization and application of intelligent transportation systems.

[0046] Example 2: 3D model optimization and texture mapping

[0047] The method consists of the following modules:

[0048] Dual-channel stereo vision data acquisition and processing module: Use the dual-channel stereo vision system to synchronously collect image data of the road surface, extract feature points in the image through image processing technology, and use the stereo matching algorithm to achieve accurate matching of feature points, providing data support for subsequent 3D reconstruction and visualization;

[0049] 3D information extraction and model reconstruction module: Based on the 3D information obtained by the dual-channel stereo vision system, a high-precision 3D model of the road surface is generated using a 3D reconstruction algorithm. Advanced algorithms and technologies are used to ensure the authenticity and accuracy of the model, providing a reliable basis for subsequent road condition analysis.

[0050] 3D model optimization and texture mapping module: The 3D model is smoothed and enhanced in detail by using a model optimization algorithm. At the same time, the model is texture mapped using image texture information to make the model more realistic, thereby improving the visual effect and authenticity of the 3D model and enabling users to more intuitively understand the actual conditions of the road surface.

[0051] Pavement condition visualization and analysis module: The present invention provides a variety of visualization methods including three-dimensional rendering and texture mapping, so that users can intuitively view the three-dimensional model of the road surface. At the same time, the module has an automatic analysis function, which can detect cracks and potholes in the road surface according to the three-dimensional model and generate corresponding inspection reports. Users can view, analyze and export the inspection reports through the interactive interface, so as to timely process and maintain the road condition.

[0052] In this embodiment, the road surface three-dimensional model previously reconstructed by stereo vision technology is optimized and texture mapped to enhance its visual effect and sense of reality. In order to improve the smoothness and accuracy of the three-dimensional model, the model is smoothed and filtered. We choose to use the Gaussian filter algorithm to smooth the model. Gaussian filtering is an algorithm widely used in image and model processing. Its mathematical expression is as follows:

[0053]

[0054] Among them, hx,y is the result after filtering, x,y are the coordinates of the point on the model surface, σ is the standard deviation of the Gaussian function, which determines the smoothness of the filter. In practical applications, the value of σ can be adjusted as needed to achieve the best smoothing effect. After Gaussian filtering, the noise and irregularities on the model surface will be effectively removed, and the smoothness and accuracy of the model will be significantly improved. In order to enhance the details and edge information of the model, the texture mapping algorithm is used to enhance the details of the model. The core formula of the texture mapping algorithm is:

[0055] Tp=LerpTv1,Tv2,α

[0056] Among them, T(p) is the texture coordinate corresponding to the point p on the model surface, Tv1 and Tv2 are the texture coordinates of the two vertices v1 and v2 adjacent to the point p, and α is the interpolation parameter, which represents the position of the point p relative to v1 and v2. Through this mapping algorithm, the texture can be smoothly fitted to the model surface, improving the realism and visual effect of the model. At the same time, combined with the 3D model optimization processing, a high-quality and detailed 3D model can be generated, providing strong support for subsequent visualization and analysis.

[0057] In summary, through the three-dimensional model optimization and texture mapping processing in this embodiment, a smooth, detailed and realistically textured three-dimensional road model can be obtained. This model can not only be used for subsequent road condition analysis and visualization, but also provide strong data support for intelligent transportation systems and autonomous driving fields.

[0058] Example 3: Application of a fast road condition detection method based on dual-path stereo vision

[0059] The specific steps of the road condition rapid detection method based on dual-path stereo vision are as follows:

[0060] (1) System preparation and initialization: Start the 3D reconstruction visualization system based on dual-channel stereo vision, ensure that the dual-channel stereo vision system operates normally, and that the camera is in a suitable position and angle to capture image data of the road surface.

[0061] (2) Dual-channel stereo vision data acquisition and processing: The dual-channel stereo vision system synchronously acquires image data of the road surface to ensure that the images captured by the left and right cameras are consistent in time for subsequent stereo matching. The acquired image data is pre-processed, including denoising and contrast enhancement, to improve image quality and prepare for subsequent feature point extraction and matching.

[0062] (3) Feature point extraction and stereo matching: In the preprocessed image, feature points are extracted using image processing technology. These feature points are usually prominent areas in the image, such as edges and corners. The feature points in the images captured by the left and right cameras are accurately matched using a stereo matching algorithm to obtain the correspondence between the feature points.

[0063] (4) 3D information extraction and model reconstruction: Based on the matching results of feature points, the 3D coordinates of the feature points are calculated through a 3D reconstruction algorithm to obtain the 3D information of the road surface. The 3D information obtained is then used to generate a high-precision 3D model of the road surface through a 3D reconstruction algorithm.

[0064] (5) 3D model optimization and texture mapping: Smooth the generated 3D model to remove noise and irregularities, improve the smoothness and realism of the model, enhance the model’s details to highlight the model’s edges and texture information, and make the model more vivid and realistic. Use the texture information of the road surface image to perform texture mapping on the 3D model, so that the model surface presents richer details and realism.

[0065] (6) Pavement condition visualization and analysis: The optimized 3D model is displayed in the form of 3D rendering, allowing users to intuitively view the 3D structure of the pavement. The system automatically analyzes the 3D model, detects cracks and potholes in the pavement, and generates corresponding inspection reports. Users can view, analyze and export the inspection reports through the interactive interface, and promptly handle and maintain the pavement condition based on the report content.

[0066] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. A method for rapid detection of road conditions based on dual-path stereo vision, characterized in that: The method consists of the following modules: Dual-channel stereo vision data acquisition and processing module: Use the dual-channel stereo vision system to synchronously collect image data of the road surface, extract feature points in the image through image processing technology, and use the stereo matching algorithm to achieve accurate matching of feature points, providing data support for subsequent 3D reconstruction and visualization; 3D information extraction and model reconstruction module: Based on the 3D information obtained by the dual-channel stereo vision system, a high-precision 3D model of the road surface is generated using a 3D reconstruction algorithm. Advanced algorithms and technologies are used to ensure the authenticity and accuracy of the model, providing a reliable basis for subsequent road condition analysis. 3D model optimization and texture mapping module: The 3D model is smoothed and enhanced in detail by using a model optimization algorithm. At the same time, the model is texture mapped using image texture information to make the model more realistic, thereby improving the visual effect and authenticity of the 3D model and enabling users to more intuitively understand the actual conditions of the road surface. Pavement condition visualization and analysis module: The present invention provides a variety of visualization methods including three-dimensional rendering and texture mapping, so that users can intuitively view the three-dimensional model of the road surface. At the same time, the module has an automatic analysis function, which can detect cracks and potholes in the road surface according to the three-dimensional model and generate corresponding inspection reports. Users can view, analyze and export the inspection reports through the interactive interface, so as to timely process and maintain the road condition.

2. The method for rapid detection of road conditions based on dual-path stereo vision according to claim 1, characterized in that: The dual-channel stereo vision data acquisition and processing module uses a dual-channel camera to synchronously capture images of the road surface from two different angles to obtain two left and right stereo images, ensure that the acquired images contain sufficient three-dimensional scene information, calibrate the dual-channel camera, obtain the internal and external parameters of the camera, such as focal length, principal point coordinates and distortion coefficient, so that in subsequent steps, the pixel points in the image can be mapped to points in three-dimensional space, pre-process the acquired stereo images, including denoising, filtering and edge detection operations to improve image quality and reduce errors in subsequent processing, extract feature points or feature lines from the pre-processed images, use a stereo matching algorithm to match the feature points or lines in the left and right images, find the corresponding points or lines in the left and right images by calculating the similarity or distance between the feature points, and thus establish a stereo correspondence.

3. The method for rapid detection of road conditions based on dual-path stereo vision according to claim 2 is characterized in that: The dual-channel stereo vision data acquisition and processing module determines the disparity value of each pixel by using the matching cost calculated by the stereo matching algorithm unit, and generates a disparity map, thereby further extracting the three-dimensional information of the road surface. The calculation formula is: C(x, y, d) = ∑ i∈N(p) |I L (x+i,y)-I R (x+i+d,y)|, where C(x, y, d) represents the matching cost of the pixel point (x, y) when the disparity is d, N(p) is the neighborhood set centered on the pixel point p(x, y), and I L (x+i,y) and I R (x+i+d,y) are the grayscale values ​​of the corresponding pixels in the left and right views respectively. Through the stereo matching algorithm, this module can accurately and efficiently process dual-channel stereo vision data to achieve high-precision three-dimensional reconstruction and visualization of the road surface.

4. The method for rapid detection of road conditions based on dual-path stereo vision according to claim 1, characterized in that: The three-dimensional information extraction and model reconstruction module calculates the disparity of each pixel point based on the result of stereo matching, combines the internal and external parameters of the camera and the disparity information, and uses the triangulation principle to calculate the coordinates of each pixel point in the three-dimensional space. Based on the calculated three-dimensional coordinate points, the three-dimensional model of the road surface is reconstructed using point cloud data or a gridding method. A smoothing algorithm is used in the reconstruction process to improve the accuracy and smoothness of the model.

5. The method for rapid detection of road conditions based on dual-path stereo vision according to claim 4 is characterized in that: The three-dimensional information extraction and model reconstruction module uses a smoothing algorithm to process the acquired three-dimensional coordinate data to reduce noise and improve model accuracy. The smoothing algorithm uses a moving average method, and its algorithm formula is: Where W is the weight, N=2m+1 is the number of moving average items, X(i) is the variable in window A, and each time window A moves forward by one data point, the variable index j also moves backward by one data point. The three-dimensional coordinate data processed by the smoothing algorithm is used in the subsequent model reconstruction steps to generate a more accurate and smoother three-dimensional model.

6. The method for rapid detection of road conditions based on dual-path stereo vision according to claim 1, characterized in that: The 3D model optimization and texture mapping module optimizes the reconstructed 3D model, including smoothing filtering and detail enhancement operations, to improve the visual effect and authenticity of the model, uses the image information of the road surface to perform texture mapping on the 3D model, and selects appropriate texture images and mapping algorithms to fit the texture information to the surface of the model, making the model more realistic and vivid.

7. The method for rapid detection of road conditions based on dual-path stereo vision according to claim 6, characterized in that: The three-dimensional model optimization and texture mapping module adopts smoothing filtering and mapping algorithms to improve the visual effect and authenticity of the model. The smoothing filtering formula is: Among them, h(x, y) is the result after filtering, (i, j) is the coordinate of the signal value to be filtered in the image, σ is the standard deviation of the Gaussian function, which determines the smoothness of the filtering. Texture mapping usually involves coordinate transformation, mapping the coordinates (u, v) in the texture space to the vertex coordinates on the surface of the three-dimensional model. Through the mapping algorithm, the noise and irregularities in the details of the model surface can be effectively removed, and the smoothness and visual effect of the model can be improved. The core formula of the texture mapping algorithm is: T(p) = Lerp(T(v1), T(v2), α), where T(p) is the texture coordinate corresponding to the point p on the model surface, T(v1) and T(v2) are the texture coordinates of the two vertices v1 and v2 adjacent to the point p, and α is the interpolation parameter, which indicates the position of the point p relative to v1 and v2. Through this mapping algorithm, the texture can be smoothly fitted to the model surface, improving the realism and visual effect of the model. At the same time, combined with the optimization processing of the three-dimensional model, a high-quality and detailed three-dimensional model can be generated, providing strong support for subsequent visualization and analysis.

8. The method for rapid detection of road conditions based on dual-path stereo vision according to claim 1, characterized in that: The pavement condition visualization and analysis module displays the optimized and texture-mapped three-dimensional model to the user in a variety of visualization methods, such as three-dimensional rendering or texture mapping. The user can zoom, rotate and roam through the interactive interface to fully and intuitively understand the actual pavement conditions. The three-dimensional model is used for automatic analysis to detect cracks or potholes in the pavement. By calculating the geometric features and color change information of the model, the location, size and severity of the disease can be identified, and a corresponding inspection report can be generated.

9. The method for rapid detection of road conditions based on dual-path stereo vision according to claim 5, characterized in that: The method also includes a calibration module, which is used to regularly calibrate the camera to ensure the accuracy and stability of the collected image data.

10. The method for rapid detection of road conditions based on dual-path stereo vision according to claim 1, characterized in that: The specific steps of the method are: (1) System preparation and initialization: Start the 3D reconstruction visualization system based on dual-channel stereo vision, ensure that the dual-channel stereo vision system operates normally, and that the camera is in a suitable position and angle to capture image data of the road surface. (2) Dual-channel stereo vision data acquisition and processing: The dual-channel stereo vision system synchronously acquires image data of the road surface to ensure that the images captured by the left and right cameras are consistent in time for subsequent stereo matching. The acquired image data is pre-processed, including denoising and contrast enhancement, to improve image quality and prepare for subsequent feature point extraction and matching. (3) Feature point extraction and stereo matching: In the preprocessed image, feature points are extracted using image processing technology. These feature points are usually prominent areas in the image, such as edges and corners. The feature points in the images captured by the left and right cameras are accurately matched using a stereo matching algorithm to obtain the correspondence between the feature points. (4) 3D information extraction and model reconstruction: Based on the matching results of feature points, the 3D coordinates of the feature points are calculated through a 3D reconstruction algorithm to obtain the 3D information of the road surface. The 3D information obtained is then used to generate a high-precision 3D model of the road surface through a 3D reconstruction algorithm. (5) 3D model optimization and texture mapping: Smooth the generated 3D model to remove noise and irregularities, improve the smoothness and realism of the model, enhance the model’s details to highlight the model’s edges and texture information, and make the model more vivid and realistic. Use the texture information of the road surface image to perform texture mapping on the 3D model, so that the model surface presents richer details and realism. (6) Pavement condition visualization and analysis: The optimized 3D model is displayed in the form of 3D rendering, allowing users to intuitively view the 3D structure of the pavement. The system automatically analyzes the 3D model, detects cracks and potholes in the pavement, and generates corresponding inspection reports. Users can view, analyze and export the inspection reports through the interactive interface, and promptly handle and maintain the pavement condition based on the report content.

Images