Road disease identification method and related device

Abstract

The application discloses a road disease identification method and related equipment, and the method comprises the following steps: acquiring a road image of a current time, and determining a plurality of sub-roads in a road area according to the road image; fitting the shape of a disease area in each sub-road, determining a hole fitting degree and a crack fitting degree of the disease area; then determining a corresponding sub-road as a disease sub-road according to the hole fitting degree and the crack fitting degree, and calculating a disease change value corresponding to the disease sub-road; finally, calculating a change index of the disease change value and a historical disease change value in a historical time to obtain a disease identification result of the road image. In the process of identifying and detecting each road, the application can identify the subtle changes in the road, avoid the fact that the road cannot be found in time due to long-time wear and tear, and can timely give a warning, thereby improving the identification accuracy of the road disease.

Description

Technical Field

[0001] This invention relates to the field of road inspection technology, specifically to a method and related equipment for identifying road defects. Background Technology

[0002] Road defects typically refer to various damages and flaws on roads. If left unrepaired for extended periods, they can increase the risk of traffic accidents. Therefore, with the development of intelligent transportation technology, road defect detection technology has emerged. This technology uses inspection robots or vehicles to collect images of designated roads, thereby determining the presence of road defects and reporting them promptly for handling. However, in this technology, each location on an urban road undergoes different changes over time. During the identification and detection process for each road segment, it is difficult to identify subtle changes, leading to insufficient accuracy in identifying road defects even after prolonged wear and tear, as these changes may persist. Summary of the Invention

[0003] Firstly, the main objective of this invention is to provide a method for identifying road defects, including:

[0004] Acquire a road image at the current time, and determine multiple sub-roads in the road area based on the road image;

[0005] Shape fitting is performed on the diseased areas in each sub-road to determine the porosity and crack fitting degree of the diseased areas;

[0006] Based on the hole fitting degree and the crack fitting degree, the corresponding sub-paths are identified as diseased sub-paths, and the disease change value corresponding to the diseased sub-paths is calculated.

[0007] The change index is calculated by comparing the change value of the road damage with the historical change value of the road damage over a historical period to obtain the road damage identification result of the road image.

[0008] Preferably, acquiring the road image at the current time and determining multiple sub-roads in the road region based on the road image includes:

[0009] When the inspection robot completes the image acquisition operation on the road, it acquires the road image at the current time;

[0010] Based on the inspection path corresponding to the road image, the direction and length information of the inspection path are determined;

[0011] The road image is divided into regions based on the direction and length information of the inspection path to determine multiple sub-roads.

[0012] Preferably, the step of performing shape fitting on the damaged areas in each sub-road to determine the porosity and crack fitting degree of the damaged areas includes:

[0013] Edge detection is performed on the diseased areas in each sub-road to determine the hole pixels and crack pixels corresponding to the diseased areas;

[0014] Based on a preset hole expansion rate, the shape of the hole pixel is fitted to obtain the corresponding hole fitted shape; and based on a preset crack expansion rate, the shape of the crack pixel is fitted to obtain the corresponding crack fitted shape.

[0015] The corresponding hole fitting degree is determined based on the pixel points corresponding to the sub-road and the pixel points corresponding to the hole fitting shape, and the corresponding crack fitting degree is determined based on the pixel points corresponding to the sub-road and the pixel points corresponding to the crack fitting shape.

[0016] Preferably, determining the corresponding sub-path as the defective sub-path based on the hole fitting degree and the crack fitting degree includes:

[0017] The average value of the defects corresponding to the sub-road is determined by weighted averaging the hole fitting degree and the crack fitting degree.

[0018] If the average value of the disease is greater than a preset threshold, the sub-road is identified as a diseased sub-road.

[0019] Preferably, calculating the disease change value corresponding to the diseased sub-road includes:

[0020] The number of holes and cracks is determined based on the diseased sub-paths;

[0021] The corresponding pore change value is determined by calculating the pore fitting degree and the number of pores, and the corresponding crack change value is determined by calculating the crack fitting degree and the number of cracks.

[0022] The disease change value corresponding to the diseased sub-road is obtained by calculating based on the change values ​​of the holes and the cracks.

[0023] Preferably, the step of calculating the change index by comparing the change value of the road damage with the historical change value of the road damage over a historical period to obtain the road damage identification result includes:

[0024] Based on the historical disease change values ​​over a historical period, multiple data collection time points corresponding to the historical disease change values ​​are determined;

[0025] Based on the current time and the multiple collection time nodes, the time interval between the current time and each collection time node is determined;

[0026] The difference between the disease change value and the historical disease change value is calculated according to each time interval, and the ratio of the obtained difference result to the corresponding time interval is calculated to determine the disease change index corresponding to each time interval.

[0027] The road image's damage identification result is obtained by summing the damage change indices corresponding to each time interval.

[0028] Preferably, the method further includes:

[0029] Pixel identification is performed on the road image to determine the pixel color corresponding to each pixel;

[0030] Classify the pixels according to their corresponding colors and identify the diseased pixel areas containing preset colors.

[0031] Based on the edge pixels of the diseased pixel region, the diseased pixel region is expanded outward, and the expanded diseased pixel region is divided into regions to obtain multiple sub-roads.

[0032] Thirdly, embodiments of the present invention provide a road defect identification device, comprising:

[0033] The acquisition module is used to acquire a road image at the current time and determine multiple sub-roads in the road area based on the road image;

[0034] The first determining module is used to perform shape fitting on the diseased area in each sub-road to determine the hole fitting degree and crack fitting degree of the diseased area.

[0035] The second determining module is used to determine the corresponding sub-path as the diseased sub-path based on the hole fitting degree and the crack fitting degree, and to calculate the disease change value corresponding to the diseased sub-path.

[0036] The calculation module is used to calculate the change index of the road damage change value and the road damage change value over a historical period to obtain the road damage identification result of the road image.

[0037] Thirdly, embodiments of the present invention provide an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the road defect identification method described above.

[0038] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the road defect identification method described above.

[0039] The above-described solution of the present invention has at least the following beneficial effects:

[0040] The road defect identification method provided by this invention first acquires a road image at the current time and determines multiple sub-roads within the road area based on the image. Then, it performs shape fitting on the defect areas within each sub-road to determine the hole fitting degree and crack fitting degree. Next, based on the hole fitting degree and crack fitting degree, the corresponding sub-road is identified as a defective sub-road, and the defect change value corresponding to the defective sub-road is calculated. Finally, the defect change value is compared with historical defect change values ​​over a historical period to calculate a change index, thus obtaining the defect identification result of the road image. This method can identify subtle changes in the road during the identification and detection process for each road segment, preventing the failure to detect road wear over a long period and enabling timely early warning, thereby improving the accuracy of road defect identification. Attached Figure Description

[0041] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the structures shown in these drawings without creative effort.

[0042] Figure 1 This is a schematic diagram of the overall process of the road defect identification method provided in an embodiment of the present invention;

[0043] Figure 2 This is a structural block diagram of the road defect identification device provided in an embodiment of the present invention;

[0044] Figure 3 This is a structural block diagram of an electronic device provided in an embodiment of the present invention.

[0045] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0046] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0047] The terms "first," "second," and "third," etc., used in the specification, claims, and accompanying drawings of this invention are used to distinguish different objects and not to describe a specific order. Furthermore, the term "comprising," and any variations thereof, is intended to cover a non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0048] First, let's take a look at the relevant accompanying drawings to illustrate the solution of the embodiments of this application.

[0049] like Figure 1 As shown, a specific embodiment of the present invention provides a method for identifying road defects, including:

[0050] 10. Obtain the road image at the current time, and determine multiple sub-roads in the road area based on the road image.

[0051] In a specific embodiment of the present invention, road images can be captured by inspection vehicles or inspection robots. The inspection robot can periodically collect corresponding road images from various road areas. The road images can be RGB images, YUV images, or HSV images, etc. The road images collected by the inspection robot can be automatically planned paths or pre-set paths. The current time can be the time when the inspection robot completes the collection of images for a certain road. Sub-roads represent the locations of road defects within a road area. Each sub-road can contain one or more road defects. Since the longer the road has been in use, the more severe the corresponding defects may be, detection and identification of each sub-road can make defect identification more accurate. Therefore, when the inspection robot completes the collection of road images for a certain road, it can upload the corresponding road images to the server for processing and identification by the server, thereby improving detection efficiency and accuracy.

[0052] Specifically, the above-mentioned acquisition of the road image at the current time and determination of multiple sub-roads in the road area based on the road image includes: acquiring the road image at the current time when the inspection robot completes the image acquisition operation of the road; determining the direction and length information of the inspection path based on the inspection path corresponding to the road image; and dividing the road image into regions based on the direction and length information of the inspection path to determine multiple sub-roads.

[0053] The aforementioned inspection path refers to the path automatically planned by the inspection robot or the path pre-set by the operator. The aforementioned direction information indicates whether the road image corresponds to a north-south or east-west direction, etc. The length information indicates the distance corresponding to the road image. Since the traffic flow is different for roads in different directions, the road image is divided into regions based on the direction information and the length information. After division, the sub-roads can be classified and filtered according to the direction information, for example, sub-roads with high traffic flow and those with low traffic flow can be classified. After classification and filtering, the images of each sub-road are processed to reduce recognition time and improve recognition efficiency. Optionally, when multiple sub-roads are identified, the images corresponding to multiple sub-roads can be enlarged and their resolution processed to make the images corresponding to the sub-roads clearer, such as performing super-resolution reconstruction on the images corresponding to the sub-roads, so as to make the recognition of the images corresponding to the sub-roads more accurate.

[0054] In an optional embodiment, the method provided in this embodiment of the invention further includes: performing pixel identification on the road image to determine the pixel color corresponding to each pixel; classifying according to the pixel color corresponding to each pixel to determine the defective pixel region containing a preset color; expanding the defective pixel region based on the edge pixels of the defective pixel region, and dividing the expanded defective pixel region into multiple sub-roads.

[0055] In a specific embodiment of the present invention, when road defects occur, the pixel color corresponding to the defect location is different from that of the road. The road image can be input into an image recognition model for pixel identification to determine the defect pixel region corresponding to the defect location. The edge pixels of the defect pixel region are generally irregular in shape, so the defect pixel region can be expanded outward to make the expansion of the defect pixel region into a rectangle. For example, the edge pixels of the defect pixel region can be expanded outward by 10 pixels in both the horizontal and vertical directions, so that each defect pixel region can be divided into a corresponding sub-road for recognition and analysis of the image corresponding to the sub-road. It can be understood that an edge detection algorithm can be used to perform edge detection on the defect pixel region to determine the corresponding edge pixels, and then divide the defect pixel region into the corresponding sub-road.

[0056] 20. Perform shape fitting on the diseased areas in each sub-road to determine the fitting degree of the holes and cracks in the diseased areas.

[0057] In embodiments of the present invention, shape fitting refers to fitting the damaged area into a geometric shape with straight or curved edges. Since the damaged shape corresponding to the damaged area will change at different times, by fitting the holes and cracks in the damaged area into corresponding geometric shapes, timely identification and early warning can be provided. It is understood that the damaged area may include holes and cracks. When performing shape fitting, the edges of holes and cracks can be basically determined based on threshold segmentation and edge detection, and then the edges of holes and cracks can be fitted. For example, it can be determined whether the edge of the hole satisfies the shape of a circle or ellipse, or whether the edge of the crack satisfies the shape of a rectangle or rhombus. Thus, the fitting degree can be calculated based on the fitted shape.

[0058] Specifically, the above-mentioned shape fitting of the diseased areas in each sub-road to determine the hole fitting degree and crack fitting degree of the diseased areas includes: performing edge detection on the diseased areas in each sub-road to determine the hole pixels and crack pixels corresponding to the diseased areas; performing shape fitting on the hole pixels based on a preset hole expansion rate to obtain the corresponding hole fitting shape, and performing shape fitting on the crack pixels based on a preset crack expansion rate to obtain the corresponding crack fitting shape; determining the corresponding hole fitting degree based on the pixels corresponding to the sub-road and the pixels corresponding to the hole fitting shape, and determining the corresponding crack fitting degree based on the pixels corresponding to the sub-road and the pixels corresponding to the crack fitting shape.

[0059] In this embodiment, the preset hole expansion rate is calculated based on the hole regions corresponding to multiple roads within a historical time period, and the preset crack expansion rate is calculated based on the crack regions corresponding to multiple roads within a historical time period. The preset hole expansion rate represents the rate at which hole pixels are fitted outward, and the preset crack expansion rate represents the rate at which crack pixels are fitted outward. For example, if a crack in a sub-road forms a straight line, the crack pixel corresponding to that crack can be fitted with a shape, which can be fitted into a rectangular shape. If a hole in a sub-road forms an irregular shape, the hole pixel corresponding to that hole can be fitted with a shape, which can be fitted into a circular shape. It can be understood that by determining the ratio between the number of pixels corresponding to the hole fitted shape and the number of pixels corresponding to the sub-road, the hole fitting degree can be determined. Similarly, by determining the ratio between the number of pixels corresponding to the crack fitted shape and the number of pixels corresponding to the sub-road, the crack fitting degree can be determined.

[0060] For example, the number of pixels corresponding to the sub-road is 2000, and the number of pixels for the hole fitting shape is 1800. Therefore, the corresponding hole fitting degree can be calculated to be 0.9. By calculating the hole fitting degree and crack fitting degree, the changing trend of holes and cracks in the sub-road can be determined, thereby timely identifying and warning of changes in the sub-road area.

[0061] 30. Based on the hole fitting degree and crack fitting degree, the corresponding sub-paths are identified as diseased sub-paths, and the disease change values ​​corresponding to the diseased sub-paths are calculated.

[0062] In a specific embodiment of the present invention, when holes and cracks are identified in a sub-road, if the size of the holes and cracks is small, they may not need to be repaired. The fitted shapes of holes and cracks are larger than their corresponding shapes. Therefore, by fitting the possible changes in the shapes of holes and cracks, the sub-road can be pre-judged using the hole fitting degree and crack fitting degree. It is understood that the defect change value represents the combined change value of the hole fitting shape and the crack fitting shape. By determining whether a sub-road is a defective sub-road, and then calculating the defect change value for that defective sub-road, the identification efficiency can be improved.

[0063] Specifically, the above-mentioned determination of the corresponding sub-road as a diseased sub-road based on the hole fitting degree and crack fitting degree includes: calculating the weighted average of the hole fitting degree and crack fitting degree to determine the average value of the disease corresponding to the sub-road; if the average value of the disease is greater than a preset threshold, the sub-road is determined as a diseased sub-road.

[0064] In this embodiment, the probability of cracks appearing in each sub-path is generally greater than the probability of holes. Therefore, the weight corresponding to the crack fitting degree can be greater than the weight corresponding to the hole fitting degree. By calculating the weighted average of the hole fitting degree and the crack fitting degree, it can be determined whether the cracks and holes in the sub-path have reached the state of needing repair. When the average defect value is greater than a preset threshold, it indicates that the cracks and holes in the sub-path need to be repaired, and thus the sub-path can be identified as a defective sub-path. Optionally, the following formula can be used to calculate the above-mentioned average defect value: w1 < w2; where K represents the average value of the disease, A represents the porosity fit, T represents the crack fit, w1 represents the weight value corresponding to the porosity fit, and w2 represents the weight value corresponding to the crack fit. By calculating the average value of the disease, the average value of the disease can be compared with a preset threshold to determine whether the sub-road is a diseased sub-road.

[0065] Furthermore, the above calculation of the disease change value corresponding to the diseased sub-road includes: determining the corresponding number of holes and cracks based on the diseased sub-road; calculating and determining the corresponding hole change value based on the hole fit degree and the number of holes, and calculating and determining the corresponding crack change value based on the crack fit degree and the number of cracks; and calculating the disease change value corresponding to the diseased sub-road based on the hole change value and the crack change value.

[0066] In this embodiment, when determining the diseased sub-path, the number of holes and cracks corresponding to the diseased sub-path can be calculated. Therefore, the product of the hole fitting degree calculated in the diseased sub-path and the corresponding number of holes can be used as the hole change value, and the product of the crack fitting degree calculated in the diseased sub-path and the corresponding number of cracks can be used as the crack change value. By summing the crack change value and the hole change value, the disease change value can be determined.

[0067] 40. Calculate the change index by comparing the change value of the road damage with the historical change value of the road damage over a historical period to obtain the road damage identification result of the road image.

[0068] In a specific embodiment of the present invention, the historical time period may include historical damage change values ​​at different time points, such as one historical damage change value per week or per day. By calculating the damage change value calculated at the current time with multiple historical damage change values ​​within the historical time period, the damage identification result corresponding to the road image can be determined. Optionally, the damage identification result can be different damage types. For example, the corresponding damage type can be determined according to different levels. When the damage change index is large, it indicates that the road damage is relatively serious, and the damage identification result corresponding to the road image can be determined as level one. When the damage change index is medium, it indicates that the road damage is relatively mild, and the damage identification result corresponding to the road image can be determined as level two. When the damage change index is small, it indicates that the road damage is relatively minor, and the damage identification result corresponding to the road image can be determined as level three. It is understood that after determining the different levels of the damage identification result, the corresponding damage identification result can be output to a visualization interface for viewing, so that staff can evaluate and repair the road after viewing to reduce the occurrence of accidents.

[0069] Specifically, the above-mentioned calculation of the change index between the change value of road defects and the historical change value of road defects within a historical period to obtain the road defect identification result includes: determining multiple collection time nodes corresponding to the historical change values ​​of road defects within a historical period; determining the time interval between the current time and each collection time node based on the current time and the multiple collection time nodes; calculating the difference between the change value of road defects and the historical change value for each time interval, and calculating the ratio of the difference result to the corresponding time interval to determine the change index of road defects for each time interval; and summing the change indices of road defects for each time interval to obtain the road defect identification result of the road image.

[0070] This process involves calculating the time interval between the current time and each data acquisition time point. Since the time interval between each data acquisition time point and the current time is different, the difference between each time interval and the corresponding difference can be calculated to determine the changes in road defects in different time intervals, making the detection more accurate. It can be understood that the defect change value calculated at the current time is used to determine the defect status of the road image at the current time. By subtracting the defect change value from the historical defect change value over a historical period, the subtle changes in road defects over a historical period can be predicted more accurately. The comprehensive calculation results are then output for staff to review.

[0071] For example, the current time is April 10th, and the corresponding disease change value is 4.1. Multiple historical data collection points are March 11th, March 22nd, and April 1st. The historical disease change value for March 11th is 2, for March 22nd it is 3.1, and for March 11th it is 3.5. Therefore, the corresponding time intervals can be calculated as 29 days, 18 days, and 10 days. By subtracting the current disease change value from the historical disease change value for each time interval, the differences are determined to be 2.1, 1, and 0.5, respectively. The corresponding change indices are calculated to be 0.07, 0.05, and 0.05, respectively. After summing these values, the corresponding disease change index is determined to be 0.17.

[0072] The road defect identification method provided by this invention first acquires a road image at the current time and identifies multiple sub-roads within the road area based on the image. Then, it performs shape fitting on the defective areas within each sub-road to determine the hole fitting degree and crack fitting degree. Next, based on the hole and crack fitting degrees, the corresponding sub-roads are identified as defective sub-roads, and the corresponding defect change value is calculated. Finally, the defect change value is compared with historical defect change values ​​over a historical period to calculate a change index, yielding the road defect identification result. This method can identify subtle changes in the road during the detection process for each segment, preventing the failure to detect road wear over a long period and enabling timely warnings, thus improving the accuracy of road defect identification.

[0073] like Figure 2 As shown, an embodiment of the present invention provides a road defect identification device 50, comprising:

[0074] The acquisition module 501 is used to acquire the road image at the current time and determine multiple sub-roads in the road area based on the road image;

[0075] The first determining module 502 is used to perform shape fitting on the diseased area in each sub-road to determine the hole fitting degree and crack fitting degree of the diseased area.

[0076] The second determining module 503 is used to determine the corresponding sub-path as the diseased sub-path based on the hole fitting degree and crack fitting degree, and to calculate the disease change value corresponding to the diseased sub-path.

[0077] The calculation module 504 is used to calculate the change index of the road damage change value and the road damage change value over a historical period to obtain the road damage identification result of the road image.

[0078] The road defect identification device 50 provided by this invention first acquires a road image at the current time and determines multiple sub-roads within the road area based on the image. It then performs shape fitting on the defect areas within each sub-road to determine the hole fitting degree and crack fitting degree. Next, based on the hole and crack fitting degrees, the corresponding sub-roads are identified as defective sub-roads, and the corresponding defect change value is calculated. Finally, the defect change value is compared with historical defect change values ​​over a historical period to calculate a change index, thus obtaining the road defect identification result. This allows for the identification and detection of subtle changes in each road segment, preventing the failure to detect road wear over a long period and enabling timely warnings, thereby improving the accuracy of road defect identification.

[0079] It should be noted that the road defect identification device 50 provided in the specific embodiments of the present invention is a device corresponding to the road defect identification method described above. All embodiments of the road defect identification method described above are applicable to the road defect identification device 50. Each embodiment of the road defect identification device 50 has a corresponding module corresponding to the steps in the road defect identification method described above, which can achieve the same or similar beneficial effects. In order to avoid excessive repetition, each module in the road defect identification device 50 will not be described in detail here.

[0080] like Figure 3 As shown, a specific embodiment of the present invention also provides an electronic device 60, including a memory 602, a processor 601, and a computer program stored in the memory 602 and executable on the processor 601. When the processor 601 executes the computer program, it implements the steps of the road defect identification method described above.

[0081] Specifically, processor 601 calls the computer program stored in memory 602 and performs the following steps:

[0082] Obtain the road image at the current time, and determine multiple sub-roads in the road area based on the road image;

[0083] Shape fitting is performed on the diseased areas in each sub-road to determine the porosity and crack fitting degree of the diseased areas;

[0084] Based on the fit degree of pores and the fit degree of cracks, the corresponding sub-roads are identified as diseased sub-roads, and the disease change value corresponding to the diseased sub-roads is calculated.

[0085] The change index is calculated by comparing the change value of the road damage with the historical change value of the road damage over a historical period to obtain the road damage identification result of the road image.

[0086] Optionally, the processor 601 performs the following steps: acquiring a road image at the current time, and determining multiple sub-roads in the road region based on the road image, including:

[0087] When the inspection robot completes the image acquisition operation on the road, it acquires the road image at the current time;

[0088] Based on the inspection path corresponding to the road image, the direction and length information of the inspection path are determined;

[0089] Based on the direction and length information of the inspection path, the road image is divided into regions to identify multiple sub-roads.

[0090] Optionally, the processor 601 performs shape fitting on the diseased areas in each sub-road to determine the porosity and crack fit of the diseased areas, including:

[0091] Edge detection is performed on the diseased areas in each sub-road to determine the corresponding hole pixels and crack pixels;

[0092] Based on the preset hole expansion rate, the shape of the hole pixel is fitted to obtain the corresponding hole fitted shape; and based on the preset crack expansion rate, the shape of the crack pixel is fitted to obtain the corresponding crack fitted shape.

[0093] The hole fitting degree is determined based on the pixel points corresponding to the sub-path and the pixel points corresponding to the hole fitting shape, and the crack fitting degree is determined based on the pixel points corresponding to the sub-path and the pixel points corresponding to the crack fitting shape.

[0094] Optionally, the processor 601 performs the following steps to determine the corresponding sub-paths as defective sub-paths based on the hole fitting degree and crack fitting degree:

[0095] The average value of the defects corresponding to the sub-road is determined by weighted average calculation based on the fit degree of the hole and the fit degree of the crack.

[0096] If the average value of the disease exceeds the preset threshold, the sub-road will be identified as a diseased sub-road.

[0097] Optionally, the processor 601 performs the calculation of the disease change values ​​corresponding to the diseased sub-roads, including:

[0098] The number of holes and cracks was determined based on the sub-paths affected by the disease.

[0099] The corresponding pore variation value is determined by calculating the pore fitting degree and the number of pores, and the corresponding crack variation value is determined by calculating the crack fitting degree and the number of cracks.

[0100] The disease change values ​​corresponding to the diseased sub-roads are calculated based on the changes in hole and crack values.

[0101] Optionally, the processor 601 performs a change index calculation by comparing the change values ​​of road defects with historical change values ​​over a historical period to obtain the road defect identification results for the road image, including:

[0102] Based on the historical disease change values ​​over a historical period, multiple data collection time points corresponding to the historical disease change values ​​were determined;

[0103] Based on the current time and multiple data collection time points, determine the time interval between the current time and each data collection time point;

[0104] The disease change value and the historical disease change value are subtracted for each time interval, and the ratio of the obtained difference result to the corresponding time interval is calculated to determine the disease change index corresponding to each time interval.

[0105] The road image disease identification results are obtained by summing the disease change index corresponding to each time interval.

[0106] Optionally, the method executed by processor 601 further includes:

[0107] Pixel identification is performed on road images to determine the color of each pixel.

[0108] Classify the pixels according to their corresponding colors and identify the diseased pixel areas containing preset colors.

[0109] Based on the edge pixels of the diseased pixel region, the diseased pixel region is expanded outward, and the expanded diseased pixel region is divided into regions to obtain multiple sub-roads.

[0110] That is, in a specific embodiment of the present invention, when the processor 601 of the electronic device 60 executes the computer program, it implements the steps of the road defect identification method described above. As a result, during the identification and detection process of each road segment, subtle changes in the road can be identified, avoiding the failure to detect road wear over a long period of time. It can also provide timely warnings and improve the accuracy of road defect identification.

[0111] It should be noted that since the processor 601 of the electronic device 60 implements the steps of the road defect identification method described above when executing the computer program, all embodiments of the road defect identification method described above are applicable to the electronic device 60 and can achieve the same or similar beneficial effects.

[0112] The computer-readable storage medium provided in this embodiment of the invention stores a computer program. When the computer program is executed by a processor, it implements the various processes of the road defect identification method or the application-side road defect identification method provided in this embodiment of the invention and can achieve the same technical effect. To avoid repetition, it will not be described again here.

[0113] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium can be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0114] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0115] The above description is only a preferred embodiment of the present invention and does not limit the patent scope of the present invention. All equivalent structural transformations made under the concept of the present invention using the contents of the present invention specification and drawings, or direct / indirect applications in other related technical fields, are included within the patent protection scope of the present invention.

Claims

1. A method for identifying road defects, characterized in that, include: Acquire a road image at the current time, and determine multiple sub-roads in the road area based on the road image; Shape fitting is performed on the damaged areas in each sub-road to determine the hole fitting degree and crack fitting degree of the damaged areas. This includes: edge detection of the damaged areas in each sub-road to determine the hole pixels and crack pixels corresponding to the damaged areas; shape fitting is performed on the hole pixels based on a preset hole expansion rate to obtain the corresponding hole fitting shape, and shape fitting is performed on the crack pixels based on a preset crack expansion rate to obtain the corresponding crack fitting shape; the corresponding hole fitting degree is determined based on the pixels corresponding to the sub-road and the pixels corresponding to the hole fitting shape, and the corresponding crack fitting degree is determined based on the pixels corresponding to the sub-road and the pixels corresponding to the crack fitting shape. Based on the hole fitting degree and the crack fitting degree, the corresponding sub-paths are identified as diseased sub-paths, and the disease change value corresponding to the diseased sub-paths is calculated, including: determining the corresponding number of holes and cracks based on the diseased sub-paths; calculating and determining the corresponding hole change value based on the hole fitting degree and the number of holes, and calculating and determining the corresponding crack change value based on the crack fitting degree and the number of cracks; and calculating the disease change value corresponding to the diseased sub-path based on the hole change value and the crack change value. The method for calculating the road image's disease identification result by comparing the disease change value with historical disease change values ​​over a historical period includes: determining multiple acquisition time nodes corresponding to the historical disease change values ​​based on the historical disease change values ​​over a historical period; determining the time interval between the current time and each of the acquisition time nodes based on the current time and the multiple acquisition time nodes; calculating the difference between the disease change value and the historical disease change value for each time interval, and calculating the ratio of the difference result to the corresponding time interval to determine the disease change index for each time interval; and summing the disease change indices for each time interval to obtain the road image's disease identification result.

2. The road defect identification method according to claim 1, characterized in that, The step of acquiring the road image at the current time and determining multiple sub-roads in the road region based on the road image includes: When the inspection robot completes the image acquisition operation on the road, it acquires the road image at the current time; Based on the inspection path corresponding to the road image, the direction and length information of the inspection path are determined; The road image is divided into regions based on the direction and length information of the inspection path to determine multiple sub-roads.

3. The road defect identification method according to claim 1, characterized in that, The step of determining the corresponding sub-path as a defective sub-path based on the hole fitting degree and the crack fitting degree includes: The average value of the defects corresponding to the sub-road is determined by weighted averaging the hole fitting degree and the crack fitting degree. If the average value of the disease is greater than a preset threshold, the sub-road is identified as a diseased sub-road.

4. The road defect identification method according to claim 1, characterized in that, The method further includes: Pixel identification is performed on the road image to determine the pixel color corresponding to each pixel; Classify the pixels according to their corresponding colors and identify the diseased pixel areas containing preset colors. Based on the edge pixels of the diseased pixel region, the diseased pixel region is expanded outward, and the expanded diseased pixel region is divided into regions to obtain multiple sub-roads.

5. A road defect identification device, characterized in that, include: The acquisition module is used to acquire a road image at the current time and determine multiple sub-roads in the road area based on the road image; The first determining module is used to perform shape fitting on the diseased areas in each sub-road to determine the hole fitting degree and crack fitting degree of the diseased areas. This includes: performing edge detection on the diseased areas in each sub-road to determine the hole pixels and crack pixels corresponding to the diseased areas; performing shape fitting on the hole pixels based on a preset hole expansion rate to obtain a corresponding hole fitting shape; and performing shape fitting on the crack pixels based on a preset crack expansion rate to obtain a corresponding crack fitting shape; determining the corresponding hole fitting degree based on the pixels corresponding to the sub-road and the pixels corresponding to the hole fitting shape; and determining the corresponding crack fitting degree based on the pixels corresponding to the sub-road and the pixels corresponding to the crack fitting shape. The second determining module is used to determine the corresponding sub-path as a diseased sub-path based on the hole fitting degree and the crack fitting degree, and to calculate the disease change value corresponding to the diseased sub-path, including: determining the corresponding number of holes and cracks based on the diseased sub-path; calculating and determining the corresponding hole change value based on the hole fitting degree and the number of holes, and calculating and determining the corresponding crack change value based on the crack fitting degree and the number of cracks; and calculating the disease change value corresponding to the diseased sub-path based on the hole change value and the crack change value. The calculation module is used to calculate the change index of the road disease change value and the change value of the road disease within a historical time period to obtain the road disease identification result. This includes: determining multiple acquisition time nodes corresponding to the historical change values ​​of the road disease within a historical time period; determining the time interval between the current time and each of the acquisition time nodes based on the current time and the multiple acquisition time nodes; calculating the difference between the change value of the road disease and the historical change value for each time interval, and calculating the ratio of the difference result to the corresponding time interval to determine the change index of the road disease for each time interval; and summing the change indices of the road disease for each time interval to obtain the road disease identification result of the road image.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the road defect identification method as described in any one of claims 1 to 4.

7. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the road defect identification method as described in any one of claims 1 to 4.

Images