Pavement disease detection method and equipment for slow road and storage medium

Abstract

The invention discloses a pavement disease detection method and device for a slow road and a storage medium, relates to the technical field of artificial intelligence, and discloses the pavement disease detection method for the slow road, and the method comprises the steps: obtaining a pavement image and pavement height data of the slow road; analyzing the road surface image and the road surface height data, and determining whether the slow road has a road surface disease or not; if the slow road has the pavement disease, constructing a first prompt text according to the pavement image, the pavement height data and the pavement disease detection requirement; the first prompt text is input into a pavement disease detection model, the pavement disease type of the slow road is obtained, the pavement disease detection model is obtained through fine adjustment according to a preset pavement disease type sample, a preset pavement image sample and a preset pavement height data sample, and therefore the accuracy of a pavement disease detection result is improved.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a method, device and storage medium for detecting road defects on a slow lane. Background Art

[0002] As an important part of the urban transportation system, the safety and comfort of slow-moving roads are increasingly valued. However, during the use of slow-moving roads, due to the dual influence of natural factors and human activities, various pavement diseases such as cracks, potholes, rutting, and peeling will occur. These diseases not only affect the appearance and service life of the road, but also pose a potential threat to the safety of pedestrians and non-motor vehicles. Therefore, timely and accurate pavement disease detection is of great significance to ensure the quality and safety of slow-moving roads. At present, the detection of pavement diseases on slow-moving roads mainly adopts the method of manual inspection. This traditional detection method relies on the field investigation of maintenance personnel, who record the disease conditions on the road surface by naked eye observation in a slow-moving vehicle or walking on the road. However, manual inspection depends largely on the experience and subjective judgment of maintenance personnel. There may be individual differences between different maintenance personnel, resulting in different judgment results for the same disease, thus affecting the accuracy of pavement disease detection results. Summary of the invention

[0003] The main purpose of the present application is to provide a method, device and storage medium for detecting pavement defects on slow lanes, aiming to solve the technical problem of low accuracy of pavement defect detection results caused by manual pavement defect detection on slow lanes.

[0004] To achieve the above purpose, the present application proposes a method for detecting road surface defects on a slow lane, comprising: Obtaining road surface images and road surface height data of the slow lane; Analyze the road surface image and road height data to determine whether there are road surface defects on the slow lane; If there are road defects on the slow lane, a first prompt text is constructed based on the road image, road height data and road defect detection requirements; The first prompt text is input into the pavement damage detection model to obtain the pavement damage type of the slow lane, wherein the pavement damage detection model is obtained by fine-tuning the preset pavement damage type sample, the preset pavement image sample and the preset pavement height data sample.

[0005] In one embodiment, analyzing the road surface image and road surface height data to determine whether there is a road surface disease on the slow lane includes: Analyze the road surface image and road surface height data to obtain road surface analysis results; According to the pavement analysis results, determine whether there are pavement defects on the slow lane; The pavement analysis results include: whether there are cracks on the pavement and whether the pavement is flat.

[0006] In one embodiment, the road surface image and the road surface height data are analyzed to obtain the road surface analysis result including: Performing image analysis on the road surface image corresponding to the target area of ​​the slow lane to determine whether there are cracks in the target area, and performing flatness analysis based on the road surface height data of the target area to determine whether the target area is flat; According to the pavement analysis results, determine whether there are pavement defects on the slow lane: If cracks exist in the target area and / or the target area is uneven, it is determined that the slow lane has pavement damage; Alternatively, if there are no cracks in the target area and the target area is flat, it is determined that there are no pavement defects on the slow lane.

[0007] In one embodiment, performing a flatness analysis based on the road height data of the target area to determine whether the target area is flat includes: Extract features from road height data to obtain height features of the target area; Based on the collection time of the road height data, the road height data is fitted to obtain the road height change trend of the target area; If the road surface height variation trend of the target area meets the preset variation trend, the target area is determined to be flat; Alternatively, if the road surface height variation trend of the target area does not conform to a preset variation trend, it is determined that the target area is uneven.

[0008] In one embodiment, the first prompt text is input into the pavement damage detection model, and the pavement damage types of the slow lane are obtained, including: Inputting the first prompt text into the pavement disease detection model, vectorizing the pavement image and pavement height data in the first prompt text, and obtaining a feature vector corresponding to the pavement image and a feature vector corresponding to the pavement height data; Comprehensively represent the feature vector corresponding to the road surface image and the feature vector corresponding to the road surface height data to obtain a comprehensive vector representation; Determine the similarity between the comprehensive vector representation and each preset vector representation in the database, wherein the database stores preset comprehensive vector representations corresponding to different road surface disease types, and the preset comprehensive vector representation is determined according to a preset feature vector corresponding to a preset road surface image and a preset feature vector corresponding to preset road surface height data; According to the similarity, the pavement damage type of the slow lane is determined.

[0009] In one embodiment, after inputting the first prompt text into the pavement damage detection model to obtain the pavement damage type of the slow lane, the method further includes: Obtain the pavement damage distribution of the slow lane; Determine a small number of sample example sets corresponding to pavement disease types and pavement disease distributions, wherein the small number of sample example sets include a plurality of preset sample examples, and the preset sample examples include preset pavement disease type samples, preset pavement disease distributions, and corresponding preset treatment suggestion samples; Constructing a second prompt text according to the pavement disease type, pavement disease distribution, a small sample example set and the processing requirements of the pavement disease type; The second prompt text is input into the pavement damage treatment model to obtain treatment suggestions for the pavement damage type and pavement damage distribution, wherein the pavement damage treatment model is fine-tuned according to preset pavement damage type samples, preset pavement damage distribution and corresponding preset treatment suggestion samples.

[0010] In one embodiment, determining a small sample set corresponding to the pavement damage type and the pavement damage distribution includes: Performing vector representation on the pavement disease type to obtain a first characteristic vector of the pavement disease type, and performing vector representation on the pavement disease distribution to obtain a second characteristic vector of the pavement disease distribution; Retrieving a first preset sample example matching the first feature vector from the sample library, and retrieving a second preset sample example matching the second feature vector; Merging the first preset sample example and the second preset sample example to obtain a merged preset sample example; The merged preset sample examples are sorted and screened by similarity to obtain a small set of sample examples.

[0011] In one embodiment, obtaining the pavement damage distribution of the slow lane includes: Obtain location data of road surface defects on slow lanes; The distribution of pavement defects is determined based on the location data of existing pavement defects.

[0012] In addition, to achieve the above-mentioned purpose, the present application also proposes a road surface disease detection device for a slow lane, the device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, the computer program being configured to implement the steps of the road surface disease detection method for a slow lane as described above.

[0013] In addition, to achieve the above-mentioned purpose, the present application also proposes a storage medium, which is a computer-readable storage medium. A computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the pavement disease detection method for slow lanes as described above are implemented.

[0014] Compared with the related art, the present application first obtains the pavement image and pavement height data of the slow lane. These two data types describe the pavement from the two dimensions of visual morphology and physical structure. Image data can capture visual features such as the color and texture of the pavement, while height data reflects the physical state of the pavement such as ups and downs and deformation. Through these two types of data, the detection process can cover various disease characteristics of the pavement more comprehensively and meticulously, reducing the detection blind spots caused by a single data type. Then, the pavement image and pavement height data are analyzed to determine whether there are pavement diseases in the slow lane, and to achieve preliminary identification of the existence of pavement diseases. Compared with manual inspection, intelligent analysis can more accurately and quickly identify the characteristics of the disease, and reduce misjudgment and missed judgment caused by human factors. When a pavement disease is detected, a prompt text containing detailed information is automatically generated according to the pavement image, height data and detection requirements. This customized text not only facilitates maintenance personnel to quickly understand the disease situation, but also provides targeted guidance for subsequent detection and repair work. The pavement disease detection model is fine-tuned based on preset pavement disease type samples, pavement image samples and pavement height data samples. This fine-tuning process enables the model to more accurately adapt to the pavement damage characteristics of specific areas and improve the accuracy of pavement damage detection results. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0017] Figure 1 A schematic diagram of a process in some embodiments of the present application; Figure 2 This is a detailed flowchart of step S20 in some embodiments of the present application; Figure 3 This is a detailed flowchart of step S211 in some embodiments of the present application; Figure 4 This is a detailed flowchart of step S40 in some embodiments of the present application; Figure 5 Another schematic diagram of a flow chart in some embodiments of the present application; Figure 6 This is a detailed flowchart of step S120 in some embodiments of the present application; Figure 7This is a schematic diagram of the structure of a road surface defect detection device for a slow lane in some embodiments of the present application.

[0018] The purpose, features and advantages of this application will be further described in conjunction with the embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0019] It should be understood that the specific embodiments described herein are only used to explain the technical solutions of the present application and are not used to limit the present application.

[0020] In order to better understand the technical solution of the present application, a detailed description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0021] At present, as an important part of the urban transportation system, the safety and comfort of slow roads are increasingly valued. However, during the use of slow roads, due to the dual influence of natural factors and human activities, various pavement diseases such as cracks, potholes, rutting, peeling, etc. will occur. These diseases not only affect the appearance and service life of the road, but also pose a potential threat to the safety of pedestrians and non-motor vehicles. Therefore, timely and accurate pavement disease detection is of great significance to ensure the quality and safety of slow roads. The detection of pavement diseases on slow roads mainly adopts the method of manual inspection. This traditional detection method relies on the field investigation of maintenance personnel. The disease conditions on the road are recorded by naked eye observation in a slow-moving vehicle or walking on the road. However, manual inspection depends largely on the experience and subjective judgment of maintenance personnel. There may be individual differences between different maintenance personnel, resulting in different judgment results for the same disease, thus affecting the accuracy of pavement disease detection results.

[0022] In response to the above problems, the present application proposes a method for detecting pavement defects on a slow lane, and the main technical solutions include: obtaining a pavement image and pavement height data of the slow lane; analyzing the pavement image and the pavement height data to determine whether there are pavement defects on the slow lane; if there are pavement defects on the slow lane, constructing a first prompt text based on the pavement image, the pavement height data and pavement defect detection requirements; inputting the first prompt text into a pavement defect detection model to obtain the pavement defect type of the slow lane, wherein the pavement disease detection model is obtained by fine-tuning the preset pavement disease type sample, the preset pavement image sample and the preset pavement height data sample.

[0023] Compared with the related art, the present application first obtains the pavement image and pavement height data of the slow lane. These two data types describe the pavement from the two dimensions of visual morphology and physical structure. Image data can capture visual features such as the color and texture of the pavement, while height data reflects the physical state of the pavement such as ups and downs and deformation. Through these two types of data, the detection process can cover various disease characteristics of the pavement more comprehensively and meticulously, reducing the detection blind spots caused by a single data type. Then, the pavement image and pavement height data are analyzed to determine whether there are pavement diseases in the slow lane, and to achieve preliminary identification of the existence of pavement diseases. Compared with manual inspection, intelligent analysis can more accurately and quickly identify the characteristics of the disease, and reduce misjudgment and missed judgment caused by human factors. When a pavement disease is detected, a prompt text containing detailed information is automatically generated according to the pavement image, height data and detection requirements. This customized text not only facilitates maintenance personnel to quickly understand the disease situation, but also provides targeted guidance for subsequent detection and repair work. The pavement disease detection model is fine-tuned based on preset pavement disease type samples, pavement image samples and pavement height data samples. This fine-tuning process enables the model to more accurately adapt to the pavement damage characteristics of specific areas and improve the accuracy of pavement damage detection results.

[0024] It should be noted that the execution subject of this embodiment is a road surface disease detection device for slow lanes that can achieve the above functions. The execution subject of this application can also be a computing service device with data processing, network communication and program running functions, such as a tablet computer, personal computer, mobile phone, etc.

[0025] In one application scenario, the pavement defect detection device for the slow lane is mobile, and can be a bicycle or an electric bicycle. The bicycle or electric bicycle is equipped with a detection module for collecting pavement images and pavement height data of the slow lane, and is also equipped with a processing module for performing pavement defect detection on the slow lane based on the collected pavement images and pavement height data.

[0026] In another application scenario, it can be connected to the above-mentioned pavement defect detection equipment for the slow lane, receive the pavement image and pavement height data collected by the pavement defect detection equipment for the slow lane, and perform pavement defect detection on the collected pavement image and pavement height data.

[0027] In another application scenario, the pavement disease detection equipment for the slow lane can collect the pavement image and pavement height data of the slow lane, and conduct a preliminary analysis of the pavement image and pavement height data of the slow lane to detect whether there is a pavement disease on the slow lane. If there is a pavement disease, a specific and accurate analysis of the pavement disease type and treatment suggestions for the pavement disease type is performed through the computing service equipment. This avoids the problem of inaccurate detection results of pavement diseases on the slow lane due to insufficient computing power of the pavement disease detection equipment for the slow lane, and can reduce the computing power of the pavement disease detection equipment for the slow lane, so that the pavement disease detection equipment for the slow lane can have more computing power to perform detection work, thereby improving the accuracy of the pavement disease detection results of the slow lane.

[0028] Based on this, the present application embodiment provides a method for detecting road surface defects in a slow lane, referring to Figure 1 , Figure 1 Schematic diagram of the flow chart of some embodiments of the method for detecting pavement defects on slow lanes of the present application.

[0029] In this embodiment, the method for detecting road surface defects on a slow lane includes steps S10 to S40: Step S10, obtaining the road surface image and road surface height data of the slow lane. Slow lanes are also called non-motorized vehicle lanes or bicycle lanes.

[0030] Road surface height data refers to the vertical distance or elevation information of each point on the surface of the slow road relative to a certain reference horizontal plane.

[0031] The road surface image and road surface height data of the slow lane may be the road surface image and road surface height data of the entire slow lane, or may be the road surface image and road surface height data of a certain target area of ​​the slow lane.

[0032] In a feasible implementation, an image acquisition sensor and a gyroscope are mounted on a pavement defect detection device for a slow lane. The pavement defect detection device for a slow lane moves along the slow lane, and the pavement image of the slow lane is acquired by the image acquisition sensor mounted thereon. At the same time, the pavement height data of the slow lane is acquired by the gyroscope mounted thereon.

[0033] In another feasible implementation, the image acquisition sensor and the gyroscope are calibrated, and then the calibrated image acquisition sensor is used to acquire the road surface image of the slow lane, and the calibrated gyroscope is used to acquire the road surface height data of the slow lane, so as to improve the accuracy of the acquired road surface image and road surface height data of the slow lane. The working parameters of the image acquisition sensor and the gyroscope can also be adjusted according to different road conditions to meet the data acquisition requirements in different road scenes.

[0034] In another feasible implementation manner, the road surface image and road surface height data of the slow lane may be collected in real time, or the road surface image and road surface height data of the slow lane may be obtained upon receiving a collection instruction.

[0035] Step S20: Analyze the road surface image and road surface height data to determine whether there are road surface defects on the slow lane.

[0036] Pavement defects refer to the various damages, deformations and other defects that will occur on cement or asphalt pavement after a period of use. Pavement defects include but are not limited to: cracks, potholes, rutting, looseness, subsidence, vehicle jumping on bridge heads and culvert tops, surface damage, etc.

[0037] In a feasible implementation, the road surface image and the road surface height data can be fused to form more complete and accurate multi-source road surface information. Among them, the road surface height data can be used to identify road surface diseases with depth information, such as potholes, subsidence, etc. The road surface image can be used to identify the surface conditions of the road surface, such as cracks, surface damage, etc. By comprehensively analyzing the multi-source road surface information, it can be more accurately determined whether there are road surface diseases.

[0038] Step S30: If there are road defects on the slow lane, a first prompt text is constructed according to the road surface image, road surface height data and road surface defect detection requirements.

[0039] The pavement disease detection requirement is proposed by maintenance personnel, and is used to guide the pavement disease detection model to achieve the purpose of identifying the pavement disease type.

[0040] The first prompt information is intended to provide sufficient context to the pavement damage detection model so that the pavement damage detection model can accurately understand the task to be performed and identify the pavement damage type accordingly. The first prompt information of the present application includes at least a pavement image, pavement height data and pavement damage detection requirements, and may also include other content.

[0041] In a feasible implementation, the pavement image, pavement height data and pavement disease detection requirements can be combined to form a first prompt text. Specifically, it can be: adding comments at the beginning or end of the pavement image and pavement height data to explain the pavement disease detection requirements, and using a specific format or template to organize the pavement image, pavement height data and pavement disease detection requirements so that the pavement disease detection model can easily parse them. In this process, the first prompt text can also be properly formatted according to the requirements of the pavement disease detection model used, which may include adjusting the pavement image, the indentation of the pavement height data, adding necessary comments or marks, etc. In addition, before the first prompt text is input into the pavement disease detection model, the first prompt text can be verified and tested, including checking the clarity of the pavement image and whether there is interference data in the pavement height data to ensure its accuracy and completeness.

[0042] Step S40, inputting the first prompt text into the pavement damage detection model to obtain the pavement damage type of the slow lane, wherein the pavement damage detection model is obtained by fine-tuning the preset pavement damage type samples, the preset pavement image samples and the preset pavement height data samples.

[0043] The pavement damage detection model is used to realize the pavement damage type of the slow lane.

[0044] The road surface disease detection model can be a large language model, which has the ability to understand natural language, can extract information from the input irregular natural language, that is, the first prompt text of this application, and actively ask questions to the user when information is missing. The core technologies of the large language model include pre-training and fine-tuning: Pre-training: During the pre-training phase, the large language model is trained using large-scale text datasets to learn the patterns, structure, and semantic information of the language.

[0045] Fine-tuning: In the fine-tuning stage, the pre-trained large language model is applied to a specific task or dataset, and the model parameters are adjusted through supervised learning to adapt it to the requirements of the specific task.

[0046] It should be noted that the pavement disease detection model used in the present application can be a general large language model or a pre-trained special large language model. It is preferably a pre-trained special large language model. The pre-trained special large language model has strong natural language processing capabilities and a certain amount of professional knowledge reserves, so that it can accurately understand the first prompt text and realize accurate identification of the pavement disease type of the slow lane.

[0047] In the embodiment of the present application, the pavement image and pavement height data of the slow lane are first obtained. These two data types describe the pavement from the two dimensions of visual morphology and physical structure. The image data can capture the visual features of the pavement such as color and texture, while the height data reflects the physical state of the pavement such as ups and downs and deformation. Through these two data, the detection process can cover various disease characteristics of the pavement more comprehensively and meticulously, reducing the detection blind spots caused by a single data type. Then, the pavement image and pavement height data are analyzed to determine whether there are pavement diseases in the slow lane, and the presence of pavement diseases is preliminarily identified. Compared with manual inspection, intelligent analysis can more accurately and quickly identify the characteristics of the disease, reducing misjudgment and missed judgment caused by human factors. When a pavement disease is detected, a prompt text containing detailed information is automatically generated according to the pavement image, height data and detection requirements. This customized text not only facilitates maintenance personnel to quickly understand the disease situation, but also provides targeted guidance for subsequent detection and repair work. The pavement disease detection model is fine-tuned based on preset pavement disease type samples, pavement image samples and pavement height data samples. This fine-tuning process enables the model to more accurately adapt to the pavement damage characteristics of specific areas and improve the accuracy of pavement damage detection results.

[0048] In some embodiments of the present application, reference Figure 2 , analyze the road surface image and road surface height data to determine whether there are road surface defects on the slow lane, including: Step S21, analyzing the road surface image and road surface height data to obtain a road surface analysis result.

[0049] The pavement analysis results include: whether there are cracks in the pavement and whether the pavement is flat, or other.

[0050] In a feasible implementation, a high-definition camera or a drone equipped with a high-definition camera can be used to collect all-round, high-resolution images of the slow lane to ensure that the image is clear and blur-free and can accurately reflect the actual condition of the road surface. The collected images are preprocessed, including denoising, contrast enhancement, color correction, etc., to improve image quality; the images are converted into formats suitable for subsequent processing, such as JPEG, PNG, etc. Image processing algorithms, such as edge detection and morphological processing, are used to automatically identify cracks in the image, and by setting appropriate thresholds and parameters, the cracks are separated from the image and their position, shape and size are marked. The road surface texture and color changes in the image are analyzed, and the flatness of the road surface is evaluated in combination with the characteristics of the road surface material. Texture analysis, grayscale co-occurrence matrix and other methods can be used to extract the texture features of the road surface and then judge the flatness of the road surface.

[0051] In another feasible implementation, high-precision measuring equipment such as laser scanners and radars can be used to collect road height data for slow lanes to ensure the accuracy and completeness of data collection and cover the entire slow lane area. The collected data is preprocessed, including denoising, filtering, interpolation, etc., to improve the accuracy and reliability of the data, and the data is converted into a format suitable for subsequent analysis, such as CSV, Excel, etc. Analyze the changes in road height data, calculate the elevation difference between adjacent points, and determine whether the road surface has diseases such as settlement and deformation based on the size and distribution of the elevation difference. Use road height data to calculate the flatness index of the road surface, such as the international flatness index, standard deviation, etc., and by comparing the flatness index with the preset threshold, determine whether the flatness of the road surface meets the requirements.

[0052] In another feasible implementation, image analysis is performed on the road surface image corresponding to a target area of ​​the slow lane to determine whether there are cracks in the target area, and a flatness analysis is performed based on the road surface height data of the target area to determine whether the target area is flat, so as to achieve the effect of identifying a target area of ​​the slow lane.

[0053] Step S22, determining whether there is any pavement damage on the slow lane based on the pavement analysis result.

[0054] In a feasible implementation, if cracks exist in the target area and / or the target area is uneven, it is determined that the slow lane has pavement damage; or, if cracks do not exist in the target area and the target area is flat, it is determined that the slow lane does not have pavement damage.

[0055] In another feasible implementation, the crack detection results in the road surface image and the flatness calculation results in the road surface height data can be combined for correlation analysis. If there are many cracks, the width is large, and the flatness index is poor, it is determined that there is a road surface disease on the slow lane. According to the shape, distribution and change of the flatness index of the cracks, the type of disease such as transverse cracks, longitudinal cracks, block cracks, etc., and the degree such as mild, moderate, and severe are determined.

[0056] The embodiment of the present application, through detailed analysis of road surface images and road surface height data, combined with methods such as crack detection and flatness assessment, can preliminarily determine whether there are road surface diseases on the slow lane, and provide a scientific basis for subsequent maintenance and repair work.

[0057] In some embodiments of the present application, reference Figure 3 , based on the road height data of the target area, the flatness analysis is performed to determine whether the target area is flat, including: Step S2111, extracting features from the road height data to obtain height features of the target area.

[0058] In a feasible implementation, the road height data of the target area is collected to ensure the integrity and accuracy of the data, and the outliers and missing values ​​in the data are cleaned to avoid interference with subsequent analysis. Key height features are extracted from the road height data, such as maximum height, minimum height, average height, height standard deviation, etc. These features can reflect the overall situation and discreteness of the road height, and provide a basis for subsequent fitting and judgment.

[0059] Step S2112, fitting the road height data based on the collection time of the road height data to obtain the road height change trend of the target area.

[0060] In a feasible implementation, the data is sorted in time series according to the collection time of the road height data, and the change trend of the time series data is analyzed, including linear trend, nonlinear trend, etc. A suitable fitting method is selected, such as linear regression, polynomial regression, exponential regression, etc., to fit the road height data. The purpose of fitting is to find a curve that can describe the change of road height over time, that is, the road height change trend line.

[0061] In another feasible implementation, after the road height variation trend line is obtained by fitting, the fitting effect of the fitting curve may be evaluated, including goodness of fit, residual analysis, etc., to ensure that the fitting curve can accurately reflect the variation trend of the road height data.

[0062] Step S2113: If the road surface height variation trend of the target area meets the preset variation trend, it is determined that the target area is flat.

[0063] The preset change trend can be set according to the actual situation. For example, according to the requirements of road surface flatness, the preset change trend can be set. The preset change trend can be flat, that is, the height change is very small, or it can have a certain slope but the change is smooth, such as a gentle slope.

[0064] Alternatively, in step S2114, if the road surface height variation trend of the target area does not conform to a preset variation trend, it is determined that the target area is uneven.

[0065] In a feasible implementation, the fitted road height change trend line is compared with the preset change trend. It is judged whether the trend line meets the requirements of the preset change trend, that is, it is judged whether the slope, curvature, etc. of the trend line are consistent with the preset value. If the road height change trend of the target area meets the preset change trend, the target area is judged to be flat. If the road height change trend of the target area does not meet the preset change trend, such as there is a sudden change, abnormal fluctuation, etc., the target area is judged to be uneven.

[0066] In another feasible implementation, the judgment result can be output in the form of a report, including the road surface height characteristics, fitting curve, judgment results, etc. of the target area, and the report should contain a detailed analysis process and judgment basis for subsequent use and reference. The judgment result can also be fed back to the relevant departments or maintenance units so that maintenance and repair work can be carried out in a timely manner. According to the feedback results and actual conditions, the plan can be improved and optimized to improve the accuracy and reliability of the judgment.

[0067] The embodiment of the present application can accurately determine whether the road surface in the target area is flat through methods such as feature extraction, data fitting, and trend comparison, and provide a scientific basis for whether there are road surface defects in the slow lane.

[0068] In some embodiments of the present application, reference Figure 4 , input the first prompt text into the pavement damage detection model, and obtain the pavement damage types of the slow lane, including: Step S41, inputting the first prompt text into the pavement defect detection model, vectorizing the pavement image and pavement height data in the first prompt text, and obtaining a feature vector corresponding to the pavement image and a feature vector corresponding to the pavement height data.

[0069] In a feasible implementation, the feature extraction module in the pavement disease detection model is used to extract features from the preprocessed pavement image to obtain a feature vector corresponding to the pavement image. Similarly, feature extraction is performed on the preprocessed pavement height data to obtain a feature vector corresponding to the pavement height data. The extracted features are converted into vector form to facilitate subsequent comprehensive representation and similarity calculation.

[0070] In another feasible implementation, an image description generation model such as Show and Tell, CaptionBot, etc. can be used to convert the road surface image into descriptive text; a simple script can be written to convert the road surface height data into text form; and the generated text description is input into the road surface disease detection model for vector representation, so that the road surface disease detection model can recognize and understand the road surface image and road surface height data. After the generated text description is input into the road surface disease detection model, the road surface disease detection model can use word embedding technology such as Word2Vec, BERT, etc. to convert the text into points in a high-dimensional vector space, in which similar text or text descriptions have similar vector representations.

[0071] Step S42, comprehensively representing the feature vector corresponding to the road surface image and the feature vector corresponding to the road surface height data to obtain a comprehensive vector representation.

[0072] In a feasible implementation, the feature vector corresponding to the road surface image and the feature vector corresponding to the road surface height data are concatenated or weighted summed to obtain a comprehensive vector representation. The comprehensive vector representation can simultaneously reflect the information of the road surface image and the road surface height data, providing a basis for subsequent similarity calculations.

[0073] Step S43, determining the similarity between the comprehensive vector representation and each preset vector representation in the database, wherein the database stores preset comprehensive vector representations corresponding to different road surface disease types, and the preset comprehensive vector representation is determined based on a preset feature vector corresponding to a preset road surface image and a preset feature vector corresponding to preset road surface height data.

[0074] In a feasible implementation, a database is established, in which preset comprehensive vector representations corresponding to different road surface disease types are stored. These preset comprehensive vector representations are obtained through the same feature extraction and vectorization process based on known preset road surface images and preset road surface height data. A suitable similarity calculation method such as cosine similarity, Euclidean distance, etc. is used to calculate the similarity between the comprehensive vector representation and each preset comprehensive vector representation in the database. The result of the similarity calculation can reflect the degree of closeness between the comprehensive vector representation and the preset comprehensive vector representation.

[0075] Step S44, determining the road surface damage type of the slow lane according to the similarity.

[0076] According to the similarity calculation result, a preset comprehensive vector representation that is most similar to the comprehensive vector representation is found, and the road surface disease type corresponding to the preset comprehensive vector representation is determined as the road surface disease type of the slow lane.

[0077] In the embodiment of the present application, the first prompt text is input into a large language model for analysis to obtain the precise road surface disease type of the slow lane.

[0078] In some embodiments of the present application, reference Figure 5 , after inputting the first prompt text into the pavement damage detection model and obtaining the pavement damage type of the slow lane, it also includes: Step S110, obtaining the pavement damage distribution of the slow lane.

[0079] The pavement damage distribution refers to the location of pavement damage on the slow lane. The pavement damage distribution can be a pavement damage distribution map, which should clearly show the location, size and severity of various types of damage on the road surface.

[0080] In a feasible implementation, while obtaining the road surface image and road surface height data of the slow lane to determine the road surface disease, the positioning data can be obtained based on the position detection sensor carried by the road surface disease detection device of the slow lane. When the road surface disease is determined to exist based on the obtained road surface image and road surface height data of the slow lane, the positioning data of the road surface disease existence in the slow lane is obtained; and then the distribution of the road surface disease is determined based on the positioning data of the road surface disease existence.

[0081] Step S120, determining a small number of sample example sets corresponding to pavement disease types and pavement disease distributions, wherein the small number of sample example sets include multiple preset sample examples, and the preset sample examples include preset pavement disease type samples, preset pavement disease distributions, and corresponding preset processing suggestion samples.

[0082] A few-shot example set is also called a few-shot example set, which means that the examples are not fixed, but similar examples are queried from the example library based on the currently input pavement disease type and pavement disease distribution. Therefore, when the pavement disease type and pavement disease distribution change, the corresponding few-shot example set also changes accordingly. Such examples are more referenceable and can further improve the accuracy of the treatment suggestions for pavement diseases. A few-shot example set can include one or more sample examples.

[0083] In a feasible implementation, a small number of representative samples are selected from existing pavement damage treatment cases as a sample set. These samples should cover different pavement damage types such as cracks, potholes, etc. and distribution patterns such as concentrated distribution and dispersed distribution.

[0084] Preset sample examples refer to pre-prepared sample examples, which are used to assist in determining the processing suggestion process. The preset sample examples include preset pavement disease type samples and preset pavement disease distributions, and the corresponding preset processing suggestion samples. Through this correspondence, the subsequent pavement disease processing model can be guided to determine the processing suggestions corresponding to the pavement disease type and pavement disease distribution. For example, the preset sample examples include pavement disease type samples such as images or descriptions of cracks, preset pavement disease distributions such as the location map of the disease on the pavement, and corresponding preset processing suggestion samples such as repair materials, repair methods, etc.

[0085] Step S130, constructing a second prompt text according to the pavement damage type, pavement damage distribution, a small sample example set and the processing requirements of the pavement damage type.

[0086] Integrate the types, distribution, and treatment requirements of the current slow-moving road pavement defects, such as urgency and budget constraints, into a descriptive text, the second prompt text. Ensure that the second prompt text has a clear structure and contains all the necessary information to facilitate the understanding and processing of the subsequent pavement defect treatment model.

[0087] Step S140, input the second prompt text into the pavement damage processing model to obtain treatment suggestions for the pavement damage type and pavement damage distribution, wherein the pavement damage processing model is fine-tuned according to preset pavement damage type samples, preset pavement damage distribution and corresponding preset treatment suggestion samples.

[0088] The pavement damage treatment model is used to recommend treatment suggestions for pavement damage on slow lanes.

[0089] The pavement disease treatment model used in this application can be a general large language model or a pre-trained special large language model, preferably a pre-trained special large language model, which has strong natural language processing capabilities and certain professional knowledge reserves, so that it can accurately understand the second prompt text and achieve accurate recommendations for treatment suggestions for pavement diseases on slow lanes. This pavement disease treatment model can also be a prediction model based on deep learning or machine learning, which can learn the relationship between disease type, distribution and treatment suggestions from the input data.

[0090] In a feasible implementation, the pavement damage treatment model outputs treatment suggestions for the current slow-moving road surface damage type and distribution according to the input second prompt text. These treatment suggestions may include specific repair methods, required materials, budget estimates, etc.

[0091] In another feasible implementation, a detailed disease treatment plan can be formulated based on the treatment suggestions output by the pavement disease treatment model. Disease treatment is carried out according to the plan, including material preparation, construction operations, etc. After the treatment is completed, the treatment effect is evaluated to ensure that the disease is effectively resolved.

[0092] In the embodiment of the present application, through the above steps, the pavement disease treatment model can intelligently provide treatment suggestions for slow-moving road pavement diseases, thereby improving the efficiency and accuracy of disease treatment.

[0093] In some embodiments of the present application, reference Figure 6 , determining the pavement damage type and the pavement damage distribution corresponding to the small number of sample example sets includes: Step S121, performing vector representation on the pavement damage type to obtain a first characteristic vector of the pavement damage type, and performing vector representation on the pavement damage distribution to obtain a second characteristic vector of the pavement damage distribution.

[0094] In a feasible implementation, for the pavement damage type, it is necessary to define a set of features that can describe the pavement damage type, such as the shape, size, color, texture, etc. of the damage. These features can be extracted from the damage image through image processing techniques such as edge detection, texture analysis, color space conversion, etc. The extracted feature values ​​are combined into a vector, namely the first feature vector of the pavement damage type. This vector can uniquely or approximately uniquely represent the damage type.

[0095] In another feasible implementation, for the distribution of pavement defects, it is necessary to extract spatial information such as the location, range, density, etc. of the defects on the pavement. This information can be obtained through geographic information systems or image processing technologies such as region segmentation and shape analysis. The extracted spatial information values ​​are combined into a vector, i.e., the second eigenvector of the pavement defect distribution, which should be able to accurately describe the distribution of the defects on the pavement.

[0096] Step S122: Retrieve a first preset sample example that matches the first feature vector from a sample library, and retrieve a second preset sample example that matches the second feature vector.

[0097] In a feasible implementation, the first feature vector of the pavement disease type is matched with a preset sample example in the sample library. The matching process can use algorithms such as vector space model, cosine similarity, and Euclidean distance to evaluate the similarity between feature vectors. The preset sample example that is most similar to the first feature vector is selected from the matching results as the first preset sample example, or the preset sample example whose similarity to the first feature vector is greater than the preset similarity is selected from the matching results as the first preset sample example.

[0098] In another feasible implementation, the second feature vector of the pavement disease distribution is matched with a preset sample example in the sample library. The preset sample example that is most similar to the second feature vector is selected from the matching results as the second preset sample example, or the preset sample example whose similarity to the second feature vector is greater than a preset similarity is selected from the matching results as the second preset sample example. The matching process can use algorithms such as vector space model, cosine similarity, and Euclidean distance to evaluate the similarity between feature vectors.

[0099] It should be noted that the first preset sample example may be one or more, and the second preset sample example may be one or more.

[0100] Step S123: merge the first preset sample example and the second preset sample example to obtain a merged preset sample example.

[0101] In a feasible implementation, the retrieved first preset sample example and the second preset sample example are merged. The merging process may involve integrating key information such as disease type description, distribution map, treatment suggestions, etc. in the two examples into a new example to obtain a merged preset sample example containing complete information.

[0102] In another feasible implementation, since the retrieval process can calculate the similarity score through the vector representation, the vector representation of each retrieved first preset sample example has a corresponding similarity score with the first feature vector. The vector representation of each retrieved second preset sample example also has a corresponding similarity score with the second feature vector. After obtaining the first preset sample example and the second preset sample example, each first preset sample example and each second preset sample example are sorted based on the corresponding similarity scores, and the top N most similar preset sample examples are selected. Wherein, N here can be 1 or an integer greater than 1.

[0103] Step S124, sorting and screening the merged preset sample examples by similarity to obtain the small number of sample example sets.

[0104] The merged preset sample examples are sorted according to the similarity evaluation results, and the examples with the highest similarity are selected as a small sample example set. The screening process can ensure that the samples in the example set are highly relevant and representative to the original input.

[0105] In a feasible implementation, the first preset sample example retrieved from the sample library is merged with the second preset sample example; if there are duplicate sample examples between the first preset sample example and the second preset sample example retrieved from the sample library, the one with the highest similarity score is retained; the merged sample examples are sorted again according to the similarity score; then the most relevant N sample examples are selected from the merged sample examples, and the most relevant N sample examples form a small sample example set. Wherein, N is an integer of 1 or greater than 1.

[0106] The embodiments of the present application can efficiently construct a small set of sample examples for specific pavement disease types and distributions, providing strong support for subsequent processing and analysis.

[0107] It should be noted that the above examples are only used to understand the present application and do not constitute a limitation on the pavement disease detection method for slow lanes of the present application. More simple transformations based on this technical concept are all within the scope of protection of the present application.

[0108] Based on the same inventive concept, the present application provides a road surface defect detection device for a slow lane, the road surface defect detection device for a slow lane comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor can execute the road surface defect detection method for the slow lane in the above-mentioned embodiment one.

[0109] Reference below Figure 7 , which shows a schematic diagram of the structure of a road surface disease detection device for a slow lane suitable for implementing an embodiment of the present application. The road surface disease detection device for a slow lane in an embodiment of the present application may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 to a random access memory (RAM: Random Access Memory) 1004. Various programs and data required for the operation of the road surface disease detection device for a slow lane are also stored in RAM1004. The processing device 1001, ROM1002 and RAM1004 are connected to each other via a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Typically, the following systems can be connected to the I / O interface 1006: input devices 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; output devices 1008 including, for example, a liquid crystal display (LCD), a speaker, a vibrator, etc.; storage devices 1003 including, for example, a magnetic tape, a hard disk, etc.; and communication devices 1009. The communication device 1009 can allow the pavement disease detection device for the slow lane to communicate with other devices wirelessly or by wire to exchange data. Although the figure shows the pavement disease detection device for the slow lane with various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or have alternatively.

[0110] In particular, according to the embodiments disclosed in the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present application include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by the processing device 1001, the above-mentioned functions defined in the method of the embodiment disclosed in the present application are executed.

[0111] The pavement disease detection device for slow lanes provided by the present application adopts the pavement disease detection method for slow lanes in the above-mentioned embodiment, which can solve the technical problem that manual pavement disease detection for slow lanes leads to low accuracy of pavement disease detection results. Compared with the prior art, the beneficial effects of the pavement disease detection device for slow lanes provided by the present application are the same as the beneficial effects of the pavement disease detection method for slow lanes provided by the above-mentioned embodiment, and the other technical features of the pavement disease detection device for slow lanes are the same as the features disclosed in the above-mentioned embodiment method, which will not be described in detail here.

[0112] It should be understood that the various parts disclosed in this application can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0113] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

[0114] Based on the same inventive concept, the present application provides a computer-readable storage medium having computer-readable program instructions (ie, computer programs) stored thereon, and the computer-readable program instructions are used to execute the pavement disease detection method for slow lanes in the above-mentioned embodiment.

[0115] The computer-readable storage medium provided in the present application may be, for example, a USB flash drive, but is not limited to electrical, magnetic, optical, electromagnetic, infrared, or semiconductor systems, systems or devices, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM: Random Access Memory), a read-only memory (ROM: Read Only Memory), an erasable programmable read-only memory (EPROM: Erasable Programmable Read Only Memory or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM: CD-Read Only Memory), an optical storage device, a magnetic storage device, or any suitable combination of the above. In this embodiment, the computer-readable storage medium may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, system or device. The program code contained on the computer-readable storage medium may be transmitted using any appropriate medium, including but not limited to: wires, optical cables, RF (Radio Frequency: Radio Frequency), etc., or any suitable combination of the above.

[0116] The computer-readable storage medium may be included in the pavement defect detection device for the slow lane; or may exist independently without being assembled into the pavement defect detection device for the slow lane.

[0117] The computer-readable storage medium carries one or more programs. When the one or more programs are executed by the pavement defect detection device for the slow lane, the pavement defect detection device for the slow lane can improve the accuracy of the pavement defect detection results.

[0118] Computer program code for performing the operations of the present application may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0119] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a module, a program segment or a part of a code, and the module, the program segment or a part of the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two square boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or flow chart, and the combination of the square boxes in the block diagram and / or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0120] The modules involved in the embodiments described in this application may be implemented by software or hardware, wherein the name of the module does not constitute a limitation on the unit itself in some cases.

[0121] The readable storage medium provided in the present application is a computer-readable storage medium, which stores computer-readable program instructions (i.e., computer programs) for executing the above-mentioned road surface disease detection method for slow lanes, and can solve the technical problem that manual road surface disease detection for slow lanes leads to low accuracy of road surface disease detection results. Compared with the prior art, the beneficial effects of the computer-readable storage medium provided in the present application are the same as the beneficial effects of the road surface disease detection method for slow lanes provided in the above-mentioned embodiments, and will not be elaborated here.

[0122] The above descriptions are only some embodiments of the present application, and are not intended to limit the patent scope of the present application. All equivalent structural changes made using the contents of the present application specification and drawings under the technical concept of the present application, or direct / indirect applications in other related technical fields are included in the patent protection scope of the present application.

Claims

1. A method for detecting road surface defects on a slow lane, characterized in that: The method for detecting road surface defects on the slow lane comprises: Obtaining road surface images and road surface height data of the slow lane; Analyze the road surface image and the road surface height data to determine whether there is road surface damage on the slow lane; If there is a road surface defect on the slow lane, construct a first prompt text according to the road surface image, the road surface height data and the road surface defect detection requirement; The first prompt text is input into a pavement damage detection model to obtain the pavement damage type of the slow lane, wherein the pavement damage detection model is obtained by fine-tuning a preset pavement damage type sample, a preset pavement image sample, and a preset pavement height data sample.

2. The method for detecting road defects on a slow lane according to claim 1, characterized in that: The analyzing the road surface image and the road surface height data to determine whether the slow lane has road surface damage includes: Analyzing the road surface image and the road surface height data to obtain a road surface analysis result; Determining whether the slow lane has pavement damage according to the pavement analysis result; The road surface analysis results include: whether there are cracks on the road surface and whether the road surface is flat.

3. The method for detecting road defects on a slow lane according to claim 2, characterized in that: The analyzing the road surface image and the road surface height data to obtain a road surface analysis result includes: Performing image analysis on a road surface image corresponding to a target area of ​​the slow lane to determine whether there are cracks in the target area, and performing flatness analysis based on road surface height data of the target area to determine whether the target area is flat; Determining whether the slow lane has pavement damage according to the pavement analysis result: If cracks exist in the target area and / or the target area is uneven, it is determined that pavement damage exists in the slow lane; Alternatively, if there are no cracks in the target area and the target area is flat, it is determined that there is no road surface damage in the slow lane.

4. The method for detecting road surface defects of a slow lane according to claim 3, characterized in that: The performing a flatness analysis based on the road surface height data of the target area to determine whether the target area is flat includes: Extracting features from the road height data to obtain height features of the target area; Fitting the road height data based on the acquisition time of the road height data to obtain a road height change trend of the target area; If the road surface height variation trend of the target area meets the preset variation trend, it is determined that the target area is flat; Alternatively, if the road surface height variation trend of the target area does not conform to the preset variation trend, it is determined that the target area is uneven.

5. The method for detecting road defects on a slow lane according to claim 1, characterized in that: The step of inputting the first prompt text into a pavement damage detection model to obtain the pavement damage type of the slow lane includes: Inputting the first prompt text into a pavement disease detection model, vectorizing the pavement image and pavement height data in the first prompt text, and obtaining a feature vector corresponding to the pavement image and a feature vector corresponding to the pavement height data; Comprehensively representing the feature vector corresponding to the road surface image and the feature vector corresponding to the road surface height data to obtain a comprehensive vector representation; Determining the similarity between the comprehensive vector representation and each preset vector representation in a database, wherein the database stores preset comprehensive vector representations corresponding to different road surface disease types, and the preset comprehensive vector representation is determined according to a preset feature vector corresponding to a preset road surface image and a preset feature vector corresponding to preset road surface height data; According to the similarity, the road surface damage type of the slow lane is determined.

6. The method for detecting road defects on a slow lane according to any one of claims 1 to 5, characterized in that: After inputting the first prompt text into a pavement damage detection model to obtain the pavement damage type of the slow lane, the method further includes: Obtaining the pavement damage distribution of the slow lane; Determine a small number of sample example sets corresponding to the pavement disease type and the pavement disease distribution, wherein the small number of sample example sets include a plurality of preset sample examples, and the preset sample examples include preset pavement disease type samples, preset pavement disease distributions, and corresponding preset treatment suggestion samples; constructing a second prompt text according to the pavement damage type, the pavement damage distribution, the small sample example set, and the processing requirements of the pavement damage type; The second prompt text is input into a pavement damage treatment model to obtain treatment suggestions for the pavement damage type and the pavement damage distribution, wherein the pavement damage treatment model is fine-tuned based on preset pavement damage type samples, preset pavement damage distribution and corresponding preset treatment suggestion samples.

7. The method for detecting road defects on a slow lane according to claim 6, characterized in that: The determining of the pavement damage type and the small sample example set corresponding to the pavement damage distribution includes: Performing vector representation on the pavement damage type to obtain a first characteristic vector of the pavement damage type, and performing vector representation on the pavement damage distribution to obtain a second characteristic vector of the pavement damage distribution; Retrieving a first preset sample example matching the first feature vector from a sample library, and retrieving a second preset sample example matching the second feature vector; Merging the first preset sample example and the second preset sample example to obtain a merged preset sample example; The merged preset sample examples are sorted and screened by similarity to obtain the small number of sample example sets.

8. The method for detecting road defects on a slow lane according to claim 6, characterized in that: The obtaining of the pavement damage distribution of the slow lane includes: Acquire location data of road surface defects on the slow lane; The distribution of pavement defects is determined based on the positioning data of the pavement defects.

9. A road surface disease detection device for a slow lane, characterized in that: The road surface disease detection device for the slow lane includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program is configured to implement the steps of the road surface disease detection method for the slow lane as described in any one of claims 1 to 8.

10. A storage medium, characterized in that: The storage medium is a computer-readable storage medium, and a computer program is stored on the storage medium. When the computer program is executed by a processor, the steps of the method for detecting pavement defects on a slow lane as described in any one of claims 1 to 8 are implemented.

Images