Road disease intelligent identification and early warning system

Abstract

The invention discloses a road disease intelligent identification and early warning system, relates to the technical field of road inspection, and solves the problems that in the image feature extraction process, road disease pixel feature analysis under a complex background is not accurate enough, and misjudgment or missed judgment of an abnormal area is easily caused. According to the method, the abnormal pixel points are accurately screened out by using the clustering algorithm through ordered analysis of the pixel points in the partition areas, and compared with a traditional method, the capability of capturing subtle disease pixel features is effectively improved; the image different region calibration end and the feature verification processing end further integrate abnormal pixel point information, and determine abnormal region feature values in combination with spatial position features, so that the spatial range and height difference of a disease region can be accurately locked, a powerful guarantee is provided for accurate recognition of diseases, and the accuracy of recognition of the diseases is improved according to the recognized specific spatial position. And the spatial position points associated with the corresponding images are locked, and the specific position features of the corresponding abnormal areas are quickly locked, so that a more accurate feature recognition result is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of road inspection, and particularly to an intelligent road disease identification and early warning system. Background Art

[0002] With the rapid development of transportation infrastructure construction, roads play an increasingly important role in the economic and social operation. However, affected by factors such as long-term vehicle loads and natural environment erosion, the problem of road diseases has become increasingly prominent. Traditional road disease detection mainly relies on manual inspections. This method not only has low efficiency, is greatly affected by the subjective judgment of inspectors, and is difficult to detect early subtle diseases; at the same time, there are certain safety hazards in manual inspections, and it cannot meet the requirements of modern transportation for the timeliness and accuracy of road maintenance.

[0003] The application with the publication number CN119808654A discloses a method and system for analyzing and early warning road and bridge diseases based on big data, which relates to the technical field of smart cities. By deploying multiple groups of sensors, the collected relevant lateral wind load information data is associated with the relevant vertical traffic load information data to construct an external force load influence coefficient Xhz. After comparison, a preliminary early warning signal is issued. Further, the relevant surface state data information and relevant vibration state data information of the current road and bridge are collected and associated with the external force load influence coefficient Xhz. Combining with a disease assessment model, the comprehensive disease degree index Zbh is fitted and calculated, and it is compared with the assessment threshold K to judge whether there is a disease risk on the road and bridge, so as to generate corresponding-level disease early warning instructions and execute them. Through the dynamic association and intelligent analysis of multi-dimensional data, the accurate identification and real-time early warning of road and bridge diseases are realized.

[0004] Some existing road disease identification technologies adopt image processing methods. Although the detection efficiency is improved to a certain extent, there are still many deficiencies. For example, in the process of image feature extraction, the pixel features of road diseases in complex backgrounds are not analyzed accurately enough, which easily leads to misjudgment or missed judgment of abnormal areas; when determining the spatial position and characteristic values of diseases, there is a lack of a systematic and scientific calculation model, and it is difficult to accurately quantify the scale and severity of diseases.

[0005] In addition, most of the existing technologies lack a complete integrated solution from image acquisition to disease early warning display, and cannot provide comprehensive and real-time decision-making support for road management departments, resulting in unreasonable allocation of road maintenance resources and untimely disease repair, affecting the service life and traffic safety of roads.

[0006] Therefore, there is an urgent need for an efficient, accurate and intelligent road disease identification and early warning system to meet the needs of modern road maintenance management. Summary of the Invention

[0007] In view of the deficiencies of the prior art, the present invention provides an intelligent road disease identification and warning system, which solves the problem that in the process of image feature extraction, the pixel features of road diseases in complex backgrounds are not analyzed accurately enough, easily leading to misjudgment or missed judgment of abnormal areas.

[0008] To achieve the above objectives, the present invention is realized through the following technical solutions: An intelligent road disease identification and warning system, comprising: An image feature processing terminal, based on the acquired relevant area images, grayscales the corresponding images, confirms the grayscale images, and then calibrates the existing abnormal pixel points based on the pixel value features associated with different points in the grayscale images. The specific method is as follows: Grayscale the acquired relevant area images, confirm the grayscale images, and then evenly divide the confirmed grayscale images to confirm several different partition areas. There is a preset equal division template during the equal division. There are several equal division cells of the same area in the equal division template, and each equal division cell corresponds to a group of partition areas; Confirm the center point of the partition area, denoted as the reference point, and then confirm the feature point farthest from the reference point within the partition area. Confirm the feature connection line between the reference point and the feature point, denote the pixel points located on the feature connection line as connection pixel points, and arrange several connection pixel points in sequence according to the relationship of being closer to the reference point from near to far to confirm a set of connection pixel point arrangements; Then process the edge pixel points within the partition area in the same way to confirm different sets of connection pixel point arrangements associated with different edge pixel points; Select the pixel points at the same sorting position from several sets of connection pixel point arrangements, confirm the pixel values associated with the corresponding several pixel points at the same sorting position, and perform feature verification on the confirmed several pixel values: Randomly select a set of pixel values as the intermediate pixel value Zh, and use [Zh - Y1, Zh + Y1] to confirm the clustering interval associated with its pixel value, where Y1 is a preset value. Take the pixel values belonging to this clustering interval as the clustering pixel values of this pixel value. Select different pixel values as the intermediate pixel value in turn, and gradually determine the clustering pixel values. Select a set of clustering intervals with the largest total number of clustering pixel values as the standard interval. Based on the confirmed standard interval, calibrate the pixel values that do not belong to this standard interval as abnormal pixel values, and calibrate the pixel points associated with the abnormal pixel values as abnormal pixel points; And calibrate the abnormal pixel points for several pixel points at different sorting positions within several sets of connection pixel point arrangements; An image abnormal area calibration terminal, for several abnormal pixel points calibrated within the relevant area images, calibrate adjacent abnormal pixel points, and confirm the specific area associated with the adjacent abnormal pixel points as the abnormal area; The feature verification processing end determines the spatial position features associated with different pixel points within the abnormal regions based on several abnormal regions calibrated in the relevant region image. Based on the confirmed spatial position features, the feature values of the abnormal regions are locked. The specific method is as follows: Confirm the spatial position of the shooting camera, confirm the vertical position of this spatial position from the confirmed relevant region image, record the pixel point associated with this vertical position as the feature verification point, record the pixel value associated with the feature verification point as Xd, and record the vertical distance associated with the feature verification point as Jl; Then calibrate the pixel values associated with different pixel points within the abnormal region as X k , where k represents different pixel points, and use: |Xd - X k | × C1 + Jl = J k Confirm the feature distance J associated with the corresponding pixel point k , where C1 is a preset fixed coefficient factor; Confirm the horizontal distance of the corresponding pixel point from the feature verification point within the relevant region image. Based on the determined horizontal distance and the location of the corresponding pixel point, calibrate this pixel point within the same plane as the feature verification point. The calibrated position is the virtual position, and then construct a set of specific perpendicular lines perpendicular to this virtual position and calibrate them as virtual perpendicular lines. Confirm the straight-line distance between different points on the virtual perpendicular line and the spatial position point of the shooting camera, and confirm the point where the straight-line distance = J k and use it as the spatial position feature point of this pixel point; Then sequentially confirm the spatial position feature points associated with different pixel points within the abnormal region; Based on the spatial position feature points associated with different pixel points within the same set of abnormal regions, confirm a set of lowest points and a set of highest points, then confirm the horizontal base plane where the lowest point is located, lock the vertical distance associated with the highest point and the horizontal base plane, and use the locked vertical distance as the feature value of this abnormal region.

[0009] Preferably, it further includes: The region image acquisition end confirms the relevant region image obtained by the image survey vehicle and transmits the confirmed relevant region image into the image feature processing end.

[0010] Preferably, it further includes: The region calibration display end confirms the diseases of different abnormal regions based on the different feature values associated with different abnormal regions, locks the disease regions and displays them. The specific method is as follows: Calibrate the different feature values associated with different abnormal regions within the relevant region image as TZ q , where q represents different abnormal regions, and satisfy: TZ qThe abnormal area with ≥Y2 is calibrated as the disease area. Conversely, no calibration is performed, and Y2 is a preset value.

[0011] The present invention provides a road disease intelligent recognition and early warning system. Compared with the prior art, it has the following beneficial effects: Through the orderly analysis of pixel points in the partitioned area, the present invention uses a clustering algorithm to accurately screen out abnormal pixel points. Compared with traditional methods, it effectively improves the ability to capture the pixel features of subtle diseases; the image different area calibration end and the feature verification processing end further integrate the abnormal pixel point information, and determine the characteristic numerical value of the abnormal area in combination with the spatial position characteristics, which can accurately lock the spatial range and height difference of the disease area, providing a strong guarantee for the accurate recognition of diseases; Based on the specific image associated with the corresponding road, the pixel features associated with the pixel points of the corresponding image are confirmed. Then, based on the pixel features associated with the corresponding points in the abnormal area, feature verification is performed to identify the spatial position where the corresponding points are located. Then, according to the specific spatial position identified, the spatial position points associated with the corresponding image are locked, and the specific position features of the corresponding abnormal area can be quickly locked, so as to achieve a more accurate feature recognition result, making the confirmation of the disease area more accurate and achieving a better recognition and early warning effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 is a schematic diagram of the principle framework of the present invention; Figure 2 is a schematic diagram showing the partitioned area of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0013] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0014] The First Embodiment Please refer to Figure 1 , this application provides a road disease intelligent recognition and early warning system, including a regional image acquisition end, an image feature processing end, an image different area calibration end, a feature verification processing end, and a regional calibration display end. Among them, the regional image acquisition end, the image feature processing end, the image different area calibration end, and the feature verification processing end are electrically connected in sequence from the output node to the input node. The eggplant feature verification processing end is electrically connected to the input node of the regional calibration display end, and the image different area calibration end is electrically connected to the input node of the regional calibration display end; Among them, the regional image acquisition end confirms the relevant regional images obtained by the image survey vehicle and transmits the confirmed relevant regional images into the image feature processing end. Specifically, generally when identifying road diseases, relevant images of the road are acquired and confirmed based on the corresponding vehicle, and then the features of the images are verified from the confirmed road-related images. Abnormal points are identified from the corresponding regional images and specifically calibrated; Among them, the image feature processing end grayscales the corresponding images based on the acquired relevant regional images, confirms the grayscale images, and then calibrates the existing abnormal pixel points based on the pixel value features associated with different points in the grayscale images. The specific method of its calibration is as follows: Grayscale the acquired relevant regional images to confirm the grayscale images (the reason for grayscaling here is to facilitate subsequent feature confirmation), and then evenly divide the area of the confirmed grayscale images to confirm several different partition regions. There is a preset equal division template for the area division. There are several equal division cells of the same area in the equal division template. Each equal division cell corresponds to a group of partition regions. The area of each partition region is the same, and when confirming the partition regions, it is confirmed according to the specific template. When the corresponding camera acquires images, there is a corresponding acquisition angle, and the height of the vehicle is in an unchanged state, so the size of the images acquired each time is the same. Therefore, according to the corresponding equal division template, the specific regional division of the images can be effectively completed; Confirm the center point of the partition region, denoted as the reference point, and then confirm the feature point that is farthest from the reference point (i.e., the straight-line distance) within the partition region. Confirm the feature connection line between the reference point and the feature point, denote the pixel points located on the feature connection line as connection pixel points, and arrange several connection pixel points in sequence according to the relationship of the distance from the reference point from near to far to confirm a set of connection pixel point arrangements; Then process the edge pixel points within the partition region in the same way to confirm different sets of connection pixel point arrangements associated with different edge pixel points (the arrangement set associated with the feature point is the longest, and the arrangement set corresponding to the edge point at the vertical distance is the shortest); Select the pixel points at the same sorting position from several sets of arranged connected pixel points, confirm the pixel values associated with the pixel points at several corresponding same sorting positions, and perform feature verification on the several confirmed pixel values: Randomly select a set of pixel values as the intermediate pixel value Zh, and use: [Zh - Y1, Zh + Y1] to confirm the clustering interval associated with its pixel value, where Y1 is a preset value, generally taking a value of 10 - 20, which is determined in advance by the operator according to experience. The pixel values belonging to this clustering interval are used as the clustering pixel values of this pixel value. Select different pixel values as the intermediate pixel value in turn, and gradually determine the clustering pixel values. Select a set of clustering intervals with the largest total number of clustering pixel values as the standard interval (for different selection processes, there are different numbers of clustering pixel values). Based on the confirmed standard interval, label the pixel values that do not belong to this standard interval as abnormal pixel values, and label the pixel points associated with the abnormal pixel values as abnormal pixel points; And label the abnormal pixel points for the pixel points at different sorting positions in several sets of arranged connected pixel points.

[0015] Specifically, combined with Figure 2 , based on the confirmed partition area, confirm the center point inside it as the corresponding reference point. After the reference point is determined, the farthest feature point and the nearest feature point associated with it can be confirmed, so that the corresponding B line and A line can be confirmed. The number of pixel points associated with the set of arranged connected pixel points on the B line is the largest, and the number of pixel points associated with the A line is the smallest; Then, based on the confirmed A line and B line, the relevant pixel points at the same sorting position can be confirmed. If there are no relevant pixel points at the corresponding sorting position in the corresponding set of pixel points, it will not be considered, and only the pixel values associated with other sets of pixel points need to be confirmed. That is to say: there are only four sets of corner points in this partition area. Then, for the pixel point at the last position in the set of arranged connected pixel points associated with the farthest feature point, there are only four pixel points at the corresponding sorting position, that is, the pixel values of the four pixel points in the last stage are subjected to verification processing.

[0016] Among them, for the image abnormal area calibration end, for several abnormal pixel points calibrated in the relevant area image, calibrate the adjacent abnormal pixel points, confirm the specific area associated with the adjacent abnormal pixel points, and record the associated specific area as the abnormal area, and calibrate it in the relevant area image; Among them, for the feature verification processing end, based on several abnormal areas calibrated in the relevant area image, determine the spatial position features associated with different pixel points in the abnormal areas. Based on the confirmed spatial position features, lock the feature values of the abnormal areas, and transmit the locked feature values to the area calibration display end. The specific method for determining the spatial position features is as follows: Confirm the spatial position of the shooting camera, confirm the vertical position of this spatial position from the relevant area images that have been confirmed, record the pixel points associated with this vertical position as feature verification points. The feature verification points are the vertical positions, and the corresponding spatial position of the shooting camera is the corresponding spatial positioning point. Combining the specific images taken, the vertical position associated with this spatial positioning point can be vertically confirmed downward, record the pixel value associated with the feature verification point as Xd, and record the vertical distance associated with the feature verification point as Jl; Then calibrate the pixel values associated with different pixel points in the abnormal area as X k , where k represents different pixel points. When the shooting object is closer to the camera, the number of pixel points it occupies in the image is relatively larger, and the actual object area corresponding to each pixel point is smaller. Therefore, it can capture the detailed information of the object more finely, and the object features reflected by the pixel values are more accurate and richer. So it can be said that the closer the distance, the relatively higher the pixel value associated with the pixel point. Use: |Xd - X k |×C1 + Jl = J k Confirm the feature distance J associated with the corresponding pixel point k , where C1 is a preset fixed coefficient factor, and its specific value is determined by the operator according to experience. Generally, C1 takes values between 5 and 10; Confirm the horizontal distance of the corresponding pixel point from the feature verification point within the relevant area images (it is impossible to confirm the distance between them within the images. Since the images are two-dimensional planes, only a horizontal distance can be confirmed). Based on the determined horizontal distance and the location of the corresponding pixel point, calibrate this pixel point within the same plane as the feature verification point. The calibrated position is the virtual position, and then construct a set of specific perpendicular lines perpendicular to this virtual position and calibrate them as virtual perpendicular lines. Confirm the straight-line distance between different positions on the virtual perpendicular line and the spatial position point of the shooting camera, and confirm the position where the straight-line distance = J k and use it as the spatial position feature point of this pixel point; Then sequentially confirm the spatial position feature points associated with different pixel points in the abnormal area; Based on the spatial position feature points associated with different pixel points within the same set of abnormal areas, confirm a set of lowest positions and a set of highest positions (by constructing a set of horizontal base planes, the height features of the corresponding spatial position feature points can be confirmed, and thus the lowest positions and the highest positions can be determined). Then confirm the horizontal base plane where the lowest position is located, lock the vertical distance associated with the highest position and the horizontal base plane, and use the locked vertical distance as the characteristic value of this abnormal area; Specifically, the so-called characteristic value refers to the height difference corresponding to the abnormal area. The value is confirmed through the correlation characteristics between the highest point and the lowest point within the corresponding abnormal area, so as to confirm the vertical distance associated between different spatial points, and thus the height difference associated with the corresponding abnormal area can be confirmed.

[0017] Among them, the area calibration display terminal confirms the diseases of different abnormal areas based on the different characteristic values associated with different abnormal areas, locks the disease areas and displays them. The specific method for disease confirmation is as follows: Calibrate the different characteristic values associated with different abnormal areas in the relevant area image as TZ q , where q represents different abnormal areas, and it satisfies: TZ q ≥Y2 of the abnormal area is calibrated as the disease area, otherwise, no calibration is performed. Y2 is a preset value, and its specific value is determined by the operator according to experience, generally taking values between 30 - 40 cm, and its specific value is determined in combination with the specific road conditions.

[0018] Some data in the above formula are numerically calculated after removing their dimensions, and the content not described in detail in this specification belongs to the prior art well-known to those skilled in the art.

[0019] The above embodiments are only used to illustrate the technical method of the present invention and not to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical method of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical method of the present invention.

Claims

1. A road disease intelligent identification and early warning system, characterized in that: include: The image feature processing end performs grayscale processing on the corresponding image based on the acquired image of the relevant area, confirms the grayscale image, and then calibrates the existing abnormal pixel points based on the pixel value features associated with different points in the grayscale image; The image different-area calibration end calibrates the adjacent abnormal pixel points for the several abnormal pixel points calibrated in the image of the relevant area, and confirms that the specific area associated with the adjacent abnormal pixel points is recorded as the abnormal area; The feature verification processing end determines the spatial position features associated with different pixel points in the abnormal area based on several abnormal areas marked in the relevant area image, and locks the feature value of the abnormal area based on the confirmed spatial position features.

2. The road hazard intelligent identification and early warning system according to claim 1 is characterized in that: Also includes: The regional image acquisition end confirms the relevant regional images acquired by the image survey vehicle and transmits the confirmed relevant regional images to the image feature processing end.

3. The road hazard intelligent identification and early warning system according to claim 1 is characterized in that: The specific method of calibrating abnormal pixels at the image feature processing end is as follows: Gray-scale the acquired image of the relevant area, confirm the gray-scale image, and then equally divide the confirmed gray-scale image into several different partitioned areas; Confirm the center point of the partition area and record it as the reference point, then confirm the feature point farthest from the reference point in the partition area, confirm the feature line between the reference point and the feature point, record the pixel points located on the feature line as the line pixel points, and arrange the plurality of line pixel points in sequence according to the relationship from near to far from the reference point to confirm a set of line pixel point arrangement sets; Then, the edge pixels in the partitioned area are processed in the same manner to determine the different line pixel arrangement sets associated with different edge pixels; Select pixel points at the same sorting position from a number of connected pixel point arrangement sets, confirm the pixel values ​​associated with the corresponding pixel points at the same sorting position, and perform feature verification on the confirmed pixel values: randomly select a group of pixel values ​​as the intermediate pixel value Zh, and use: [Zh-Y1, Zh+Y1] to confirm the clustering interval associated with its pixel value, where Y1 is a preset value, and use the clustering pixel value belonging to this clustering interval as the clustering pixel value of this pixel value, select different pixel values ​​as the intermediate pixel values ​​in turn, and gradually determine the clustering pixel values, and select a group of clustering intervals with the largest total number of clustering pixel values ​​as the standard interval, based on the confirmed standard interval, mark the pixel values ​​that do not belong to this standard interval as abnormal pixel values, and mark the pixel points associated with the abnormal pixel values ​​as abnormal pixel points; And the abnormal pixel points are calibrated for several pixel points at different sorting positions in several connected pixel point arrangement sets.

4. The road hazard intelligent identification and early warning system according to claim 3 is characterized in that: When the area is equally divided, there is a preset equally divided template, and there are a number of equally divided grids with the same area in the equally divided template, and each equally divided grid corresponds to a group of partitioned areas.

5. The road hazard intelligent identification and early warning system according to claim 1 is characterized in that: The specific method of determining the spatial position features of different pixels at the feature verification processing end is as follows: Confirm the spatial position of the shooting camera, confirm the vertical point of this spatial position from the confirmed relevant area image, record the pixel point associated with this vertical point as a feature check point, record the pixel value associated with the feature check point as Xd, and record the vertical distance associated with the feature check point as Jl; Then the pixel values ​​associated with different pixel points in the abnormal area are calibrated as X k , where k represents different pixels, using: |Xd-X k |×C1+Jl=J k Confirm the feature distance J associated with the corresponding pixel point k , where C1 is the preset fixed coefficient factor; Confirm the horizontal distance between the corresponding pixel point and the feature check point in the image of the relevant area. Based on the determined horizontal distance and the location of the corresponding pixel point, calibrate the pixel point in the same plane as the feature check point. The calibrated position is the virtual position. Then, construct a set of specific vertical lines perpendicular to the virtual position and calibrate them as virtual vertical lines. Confirm the straight-line distance between different points and the spatial position of the shooting camera from the virtual vertical lines. Confirm the straight-line distance = J k The point is taken as the spatial position feature point of this pixel; Then, the spatial position feature points associated with different pixel points in the abnormal area are confirmed in turn; Based on the spatial position feature points associated with different pixel points in the same group of abnormal areas, a group of lowest points and a group of highest points are confirmed, and then the horizontal base plane where the lowest points are located is confirmed, and the vertical distance associated with the highest point and the horizontal base plane is locked, and the locked vertical distance is used as the characteristic value of this abnormal area.

6. The road hazard intelligent identification and early warning system according to claim 1 is characterized in that: Also includes: The area calibration display terminal confirms the disease of different abnormal areas based on the different characteristic values ​​associated with different abnormal areas, locks the diseased areas and displays them.

7. The road hazard intelligent identification and early warning system according to claim 6, characterized in that: The specific method of locking the diseased area at the area calibration display terminal is as follows: The different feature values ​​associated with different abnormal areas in the relevant area image are calibrated as TZ q , where q represents different abnormal regions, will satisfy: TZ q The abnormal area with value ≥Y2 is marked as the diseased area.

8. The road hazard intelligent identification and early warning system according to claim 7, characterized in that: Will not satisfy: TZ q For abnormal areas ≥Y2, no calibration is performed and Y2 is the preset value.

Images