Road crack monitoring and early warning method and system based on machine vision

Abstract

The invention relates to the technical field of road monitoring and early warning, and particularly discloses a road crack monitoring and early warning method and system based on machine vision, and the system comprises an image collection module which comprises a camera device disposed at the top of a vehicle and is used for collecting road condition information in front of the vehicle according to a fixed time interval, the road condition information in front of the automobile at the current time point can be reflected by analyzing the road crack information and the state information of the automobile in the driving process through the road evaluation module, and the accuracy of the road condition information evaluation result can be improved based on diversified data analysis; then, the driving risk of the vehicle on the current road is analyzed and judged in combination with the data of the vehicle, and the driving risks of different types of vehicles under the road condition can be evaluated, so that the accuracy of the driving risk judgment result is ensured, wrong early warning information is prevented from being output to a driver, and the driving safety is improved. The judgment of the driver is influenced, and the risk of traffic accidents is increased.

Description

Technical Field

[0001] The present invention relates to the technical field of road monitoring and early warning, and in particular to a road crack monitoring and early warning method and system based on machine vision. Background Art

[0002] Road crack monitoring and early warning are important parts of road maintenance and safety management. An effective monitoring and early warning system can detect road crack problems in a timely manner and issue early warnings to drivers to ensure driving safety.

[0003] The traditional method of monitoring road cracks is to use a high-resolution camera in the vehicle to capture image information of the road surface, and to process and analyze the images captured by the camera in real time to identify the location, size, shape and other characteristics of the cracks. The driving risk of the car will then be judged based on the identified information. When the driving risk is judged to be high, an early warning will be issued to remind the driver to avoid affecting the driver's control and judgment and increasing the risk of traffic accidents.

[0004] The common method in the prior art is to judge the driving risk by combining the image information of the road surface. The data only reflects the damage of the road. However, the driving conditions of different types of vehicles on cracked roads are different. Therefore, the accuracy of the judgment result of driving risk based on the road condition information alone is low, which may output wrong warning information to the driver, thereby affecting the driver's judgment and increasing the risk of traffic accidents. Summary of the invention

[0005] The purpose of the present invention is to provide a road crack monitoring and early warning method and system based on machine vision to solve the following technical problems:

[0006] How to improve the accuracy of vehicle driving risk judgment results.

[0007] The purpose of the present invention can be achieved through the following technical solutions:

[0008] A road crack monitoring and early warning method and system based on machine vision, the system comprising:

[0009] An image acquisition module, including a camera device disposed on the top of the vehicle, for acquiring road condition information in front of the vehicle at fixed time intervals;

[0010] An image recognition module is used to recognize the image information collected by the image collection module and extract the road crack information in the image;

[0011] The data recording module is used to collect and record the vehicle's status information and environmental data during driving;

[0012] The road assessment module is used to combine the information obtained by the image recognition module with the vehicle's state information during driving, calculate the abnormality index of the road ahead of the vehicle at different time points, and judge whether there is an abnormality on the current road based on the preset road abnormality index threshold;

[0013] The risk analysis module is used to analyze and judge the driving risk of the vehicle on the current road by combining the vehicle's own data with the evaluation results of the road evaluation module;

[0014] The early warning module is used to combine the analysis results of the risk analysis module and the road assessment module to decide whether an early warning is needed.

[0015] Furthermore, the road crack information identified by the image recognition module includes: road damage area, road damage length and maximum road damage depth, and the data collected by the data recording module includes: vehicle speed, vibration amplitude, light intensity in the current shooting environment and dust content in the air during driving.

[0016] Furthermore, the evaluation process of the road evaluation module includes:

[0017] By formula Calculate the image loss coefficient δ at the i-th time point i ;

[0018] Among them, i is a data collection at a fixed time interval, zd i is the vehicle vibration amplitude at the i-th time point, zd y is the preset vibration amplitude, sd i is the vehicle speed at the i-th time point, sd b sd i The standard value of μ is the error correction coefficient. According to the setting of empirical fitting, gx i is the light intensity at the ith time point, gx y is the preset light intensity, gx b For gx i The standard value of c ai is the dust content in the air at the i-th time point, ca b c ai The standard value of , x1 and x2 are the first weight coefficients.

[0019] Furthermore, the evaluation process of the road evaluation module also includes:

[0020] By formula Calculate the abnormal index ρ of the road ahead of the vehicle at the i-th time point i ;

[0021] Where a is any damaged area on the road ahead of the vehicle at the i-th time point, b is the total number of damaged areas on the road ahead of the vehicle at the i-th time point, and pos ai is the road damage area of ​​the ath damaged area in the road ahead of the vehicle at the i-th time point, z a is the weight coefficient of the a-th damaged area, which is set according to empirical fitting, γ a is the damage position influence coefficient of a damaged area, which is set according to empirical fitting, pos iz is the total area of ​​the road ahead of the vehicle at the i-th time point, sd ai is the road damage length of the ath damaged area in the road ahead of the vehicle at the i-th time point, sd y is the preset road damage length, cd i is the maximum depth of road damage in front of the vehicle at the i-th time point, cd y is the preset maximum depth of road damage, f t To adjust the coefficient lookup table function, according to the influence of the value range of the image loss coefficient at the i-th time point on the abnormality index of the road ahead of the vehicle, y1 and y2 are the second weight coefficients based on the test.

[0022] Furthermore, the evaluation process of the road evaluation module also includes:

[0023] By calculating the abnormal index ρ of the road ahead of the vehicle at the i-th time point i and the preset road abnormality coefficient threshold ρ 01 Make a comparison;

[0024] If i ≤ρ 01 , the system determines that the abnormality index of the road ahead of the vehicle at the i-th time point is low, which means that the road condition ahead of the vehicle at the i-th time point is good, and no warning is issued;

[0025] If i >ρ 01 The system determines that the abnormal index of the road ahead of the vehicle at the i-th time point is high, which means that the road condition ahead of the vehicle at the i-th time point is abnormal. The system issues an early warning through the early warning module and conducts risk analysis in combination with the vehicle's own data.

[0026] Furthermore, the analysis process of the risk analysis module includes:

[0027] By formula Calculate the risk index of the car driving at the i-th time point

[0028] Among them, dp is the chassis height of the car, dp b is the standard value of dp, xg is the suspension strength of the car, yis the preset suspension strength, lts is the usage time of the vehicle tire, lts y is the preset usage time of the tire, ω is the tire pressure influence coefficient, which is set based on empirical fitting, and jscl is the driver's driving experience.

[0029] Furthermore, the analysis process of the risk analysis module also includes:

[0030] By calculating the risk index of the car driving at the i-th time point Compared with the preset risk index threshold Make a comparison;

[0031] like The system determines that the risk of the car driving at the i-th time point is low, and the vehicle is not prone to bumps and vibrations when driving, which does not affect the driver's control and judgment;

[0032] like The system determines that the car driving risk is high at the i-th time point. The vehicle is prone to strong bumps and vibrations when driving, which affects the driver's control and judgment, and issues an early warning through the early warning module to remind the driver to slow down.

[0033] A road crack monitoring and early warning method based on machine vision, the method comprising:

[0034] S1: collecting road condition information in front of the vehicle at fixed time intervals through an image acquisition module arranged on the top of the vehicle;

[0035] S2: The image information collected by the image acquisition module is recognized by the image recognition module, and the road crack information in the image is extracted;

[0036] S3: Collecting the vehicle status information and environmental data during driving through the data recording module and recording them;

[0037] S4: The road evaluation module calculates the abnormality index of the road ahead of the vehicle at different time points, and judges whether there is an abnormality on the current road in combination with the preset road abnormality index threshold, and the early warning module combines the analysis results to decide whether an early warning is needed;

[0038] S5: The risk analysis module combines the vehicle's own data with the evaluation results of the road evaluation module to analyze and judge the driving risk of the vehicle on the current road, and the early warning module combines the analysis results to decide whether an early warning is needed.

[0039] Beneficial effects of the present invention:

[0040] (1) The present invention combines the road crack information with the vehicle status information during driving for analysis through a road assessment module, which can reflect the road condition information in front of the vehicle at the current time point, and based on diversified data analysis, can improve the accuracy of the road condition information assessment results. Then, by combining the vehicle's own data to analyze and judge the vehicle's driving risk on the current road, the driving risk of different models of vehicles under the road condition can be assessed, thereby ensuring the accuracy of the driving risk judgment result and avoiding the output of erroneous warning information to the driver, which affects the driver's judgment and increases the risk of traffic accidents.

[0041] (2) The present invention improves the accuracy of the calculation result by correcting the road anomaly index ahead of the vehicle at the ith time point in combination with the road anomaly index ahead of the vehicle at the ith time point. Moreover, since the data can intuitively reflect the road condition ahead of the vehicle at the current time point, the accuracy of the road condition information can be further improved through this calculation method, thereby providing accurate data for subsequent judgment of road anomalies and ensuring the accuracy of the judgment result.

[0042] (3) The present invention calculates the road anomaly index ρ ahead of the vehicle at the i-th time point i and the preset road abnormality coefficient threshold ρ 01 By comparing, through this comparison method, the road condition ahead of the vehicle at the ith time point can be analyzed according to the height of the abnormality index of the road ahead of the vehicle at the ith time point, and an early warning can be issued to remind the driver to concentrate when it is judged that the road condition is abnormal. The road condition information can make a preliminary judgment on the driving risk by monitoring the cracks in the road, and then by combining the vehicle's own data for risk analysis, the accuracy of the vehicle's driving risk judgment results can be improved.

[0043] (4) The present invention calculates the risk index of the vehicle driving at the i-th time point Compared with the preset risk index threshold By comparing, an accurate judgment can be made on the driving risk of the car at the i-th time point, and when it is judged that the driving risk of the car at the i-th time point is high, an early warning module is used to issue an early warning in time to remind the driver to reduce the speed. By setting it in this way, after combining vehicle information data and diversified data such as road cracks, the accuracy of the judgment results of the car driving risk can be improved, thereby ensuring the output of correct early warning information to the driver to avoid affecting the driver's judgment.

[0044] (5) The image acquisition module of the present invention monitors road cracks and analyzes the road crack information in combination with the vehicle status information during driving, so as to make an accurate judgment on the road condition information in front of the vehicle. Then, by combining the data with the vehicle's own data, the driving risk of the vehicle can be further judged. The diversified data are linked together, thereby improving the accuracy of the judgment result, thereby ensuring that an early warning can be issued in time to remind the driver in the event of road abnormalities or high driving risks, thereby improving the driver's vigilance and preventing the occurrence of traffic accidents. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] The present invention will be further described below in conjunction with the accompanying drawings.

[0046] Figure 1 It is a schematic block diagram of a road crack monitoring and early warning system based on machine vision in the present invention;

[0047] Figure 2 It is a flow chart of a road crack monitoring and early warning method based on machine vision in the present invention. DETAILED DESCRIPTION

[0048] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0049] See also Figure 1 As shown, in one embodiment, the present application provides a road crack monitoring and early warning method and system based on machine vision, the system comprising:

[0050] An image acquisition module, including a camera device disposed on the top of the vehicle, for acquiring road condition information in front of the vehicle at fixed time intervals;

[0051] An image recognition module is used to recognize the image information collected by the image collection module and extract the road crack information in the image;

[0052] The data recording module is used to collect and record the vehicle's status information and environmental data during driving;

[0053] The road assessment module is used to combine the information obtained by the image recognition module with the vehicle's state information during driving, calculate the abnormality index of the road ahead of the vehicle at different time points, and judge whether there is an abnormality on the current road based on the preset road abnormality index threshold;

[0054] The risk analysis module is used to analyze and judge the driving risk of the vehicle on the current road by combining the vehicle's own data with the evaluation results of the road evaluation module;

[0055] The early warning module is used to combine the analysis results of the risk analysis module and the road assessment module to decide whether an early warning is needed;

[0056] Through the above technical solution, this embodiment provides an image acquisition module, which is arranged on the top of the vehicle. During the driving process of the vehicle, the image acquisition module will collect the road condition information in front of the vehicle at fixed time intervals, and then identify the image information collected by the image acquisition module to extract the road crack information in the image, and collect the state information and environmental data of the vehicle during the driving process through the data recording module. After that, the road evaluation module combines the information obtained by the image recognition module and the state information of the vehicle during the driving process to calculate the abnormal index of the road in front of the vehicle at different time points, and combines the preset road abnormality index threshold to make a judgment on whether there is an abnormality on the current road. Finally, the risk analysis module can combine the vehicle's own data with the evaluation result of the road evaluation module to analyze and judge the driving risk of the vehicle on the current road;

[0057] Through such a setting, during the driving process of the vehicle, the road assessment module combines the road crack information with the vehicle status information during the driving process for analysis, which can reflect the road condition information in front of the car at the current time point, and based on the diversified data analysis, the accuracy of the road condition information assessment result can be improved. After that, the driving risk of the vehicle on the current road is analyzed and judged by combining the vehicle's own data, and the driving risk of different models of vehicles under the road condition can be assessed, thereby ensuring the accuracy of the driving risk judgment result, avoiding the output of wrong warning information to the driver, resulting in affecting the driver's judgment and increasing the risk of traffic accidents;

[0058] In addition, the early warning module in this system can monitor the analysis results of the risk analysis module and the road assessment module, and decide whether an early warning is needed based on the monitoring results, thereby ensuring that early warnings are given in time to remind the driver in the event of road abnormalities or high driving risks, thereby increasing the driver's vigilance and preventing traffic accidents.

[0059] The road crack information identified by the image recognition module includes: road damage area, road damage length and maximum road damage depth; the data collected by the data recording module includes: vehicle speed, vibration amplitude, light intensity in the current shooting environment and dust content in the air during driving;

[0060] Through the above technical scheme, this embodiment provides road crack information recognized by the image recognition module and data collected by the data recording module, wherein the road crack information recognized by the image recognition module can intuitively reflect the road conditions of the road ahead when the vehicle is currently traveling, while the data collected by the data recording module reflects the vehicle's driving status and current environmental information. This data may cause deviations in the image data collected by the image acquisition module, so by collecting this data, the image data collected by the image acquisition module can be corrected and cleaned, thereby improving the accuracy of the road crack information recognized by the subsequent image recognition module, and accurate data can improve the accuracy of the subsequent driving risk judgment results.

[0061] The evaluation process of the road evaluation module includes:

[0062] By formula Calculate the image loss coefficient δ at the i-th time point i ;

[0063] Among them, i is a data collection at a fixed time interval, zd i is the vehicle vibration amplitude at the i-th time point, zd y is the preset vibration amplitude, sd i is the vehicle speed at the i-th time point, sd b sd i The standard value can be set according to the allowable error in the empirical data. μ is the error correction coefficient. According to the setting of empirical fitting, gx i is the light intensity at the ith time point, gx y is the preset light intensity, gx b For gx i The standard value can be selected and set according to the allowable error in the empirical data, ca i is the dust content in the air at the i-th time point, ca b c ai The standard value of , which can be selected and set according to the allowable error in the empirical data, x1 and x2 are the first weight coefficients, which can be set according to empirical fitting;

[0064] Through the above technical solution, this embodiment provides the image loss coefficient δ at the i-th time point i , can be obtained by formula Obviously, when the vehicle vibration amplitude, vehicle speed and dust content in the air at the i-th time point are higher, and the difference between the light intensity at the i-th time point and the preset light intensity is larger, the image loss coefficient δ at the i-th time point is iThe larger the value is, the higher the vehicle vibration amplitude and vehicle speed at the i-th time point are, the worse the stability of the camera device in the image acquisition module is, and the lower the quality of the image it captures is. The higher the dust content in the air is, the more blurred the image will be. In addition, too high or too low light intensity will affect the quality of the captured image.

[0065] Therefore, when the vehicle vibration amplitude, vehicle speed and dust content in the air at the i-th time point are lower, and the difference between the light intensity at the i-th time point and the preset light intensity is smaller, the image loss coefficient δ at the i-th time point is i The smaller it is, the clearer the captured image is. Through this calculation method, the data can be combined to correct and clean the quality of the image taken by the camera device, making it closer to the real data, thereby improving the authenticity and accuracy of the data.

[0066] The evaluation process of the road evaluation module also includes:

[0067] By formula Calculate the abnormal index ρ of the road ahead of the vehicle at the i-th time point i ;

[0068] Where a is any damaged area on the road ahead of the vehicle at the i-th time point, b is the total number of damaged areas on the road ahead of the vehicle at the i-th time point, and pos ai is the road damage area of ​​the ath damaged area in the road ahead of the vehicle at the i-th time point, z a is the weight coefficient of the a-th damaged area, which is set according to empirical fitting, γ a is the damage position influence coefficient of a damaged area, which is set according to empirical fitting, pos iz is the total area of ​​the road ahead of the vehicle at the i-th time point, sd ai is the road damage length of the ath damaged area in the road ahead of the vehicle at the i-th time point, sd y is the preset road damage length, cd i is the maximum depth of road damage in front of the vehicle at the i-th time point, cd y is the preset maximum depth of road damage, f t To adjust the coefficient comparison table function, according to the influence of the value range of the image loss coefficient at the i-th time point on the abnormal index of the road ahead of the vehicle, based on the test, y1 and y2 are the second weight coefficients, which are set according to empirical fitting;

[0069] Through the above technical solution, this embodiment provides the vehicle front road abnormality index ρ at the i-th time point i , can be obtained by formula Obviously, the larger the road damage area and road damage length of the ath damaged area in the road ahead of the vehicle at the i-th time point, and the higher the maximum depth of the road damage, the higher the abnormal index ρ of the road ahead of the vehicle at the i-th time point i The higher the index, the worse the road condition of the road ahead of the vehicle at the current time point. Conversely, when the road damage area and road damage length of the ath damaged area in the road ahead of the vehicle at the i-th time point are smaller and the maximum depth of the road damage is lower, then the abnormality index ρ of the road ahead of the vehicle at the i-th time point is i The lower the index, the better the road condition ahead of the vehicle at the current time.

[0070] By correcting the abnormality index of the road ahead of the vehicle at the ith time point in combination with the abnormality index of the road ahead of the vehicle at the ith time point, the accuracy of the calculation result is improved, and because the data can intuitively reflect the road condition in front of the vehicle at the current time point, this calculation method can further improve the accuracy of the road condition information, thereby providing accurate data for subsequent judgments on road abnormalities and ensuring the accuracy of the judgment results.

[0071] The evaluation process of the road evaluation module also includes:

[0072] By calculating the abnormal index ρ of the road ahead of the vehicle at the i-th time point i and the preset road abnormality coefficient threshold ρ 01 Make a comparison;

[0073] If i ≤ρ 01 , the system determines that the abnormality index of the road ahead of the vehicle at the i-th time point is low, which means that the road condition ahead of the vehicle at the i-th time point is good, and no warning is issued;

[0074] If i >ρ 01 , the system determines that the abnormality index of the road ahead of the vehicle at the i-th time point is high, which means that the road condition ahead of the vehicle at the i-th time point is abnormal, and issues an early warning through the early warning module, and performs risk analysis in combination with the vehicle's own data;

[0075] Through the above technical solution, this embodiment calculates the abnormality index ρ of the road ahead of the vehicle at the i-th time point i and the preset road abnormality coefficient threshold ρ 01By comparing, through this comparison method, the road condition ahead of the vehicle at the ith time point can be analyzed according to the height of the abnormality index of the road ahead of the vehicle at the ith time point, and an early warning can be issued to remind the driver to concentrate when it is judged that the road condition is abnormal. The road condition information can make a preliminary judgment on the driving risk by monitoring the cracks in the road, and then by combining the vehicle's own data for risk analysis, the accuracy of the vehicle's driving risk judgment results can be improved.

[0076] The analysis process of the risk analysis module includes:

[0077] By formula Calculate the risk index of the car driving at the i-th time point

[0078] Among them, dp is the chassis height of the car, dp b is the standard value of dp, which can be set based on the allowable error in empirical data, xg is the suspension strength of the car, xg y is the preset suspension strength, lts is the usage time of the vehicle tire, lts y is the preset usage time of the tire, ω is the tire pressure influence coefficient, which is set according to experience fitting, jscl is the driver's driving age, and g is the proportional coefficient, which is set according to experience fitting;

[0079] Through the above technical solution, this embodiment provides the risk index of automobile driving at the i-th time point: The formula The calculation shows that the higher the chassis of the car is, the stronger its passability is. The older the driver is, the richer his driving experience is. The lower the tire usage time and suspension strength are, the less tire wear the vehicle has. On the other hand, the better the shock absorption effect of the vehicle is. Therefore, on this basis, when the chassis height of the car and the driver's driving age are higher, and the tire usage time and suspension strength are lower, the risk index of the car driving at the i-th time point is On the contrary, when the chassis height of the car and the driver's driving experience are lower, and the tire usage time and suspension strength are higher, the risk index of the car driving at the i-th time point is the higher;

[0080] Through this calculation method, the vehicle data of different vehicles and the driver's conditions can be combined to further calculate the driving risk. The diversified data can improve the accuracy of the calculation results. Based on accurate data information, the system can make accurate judgments on the level of subsequent car driving risks.

[0081] The analysis process of the risk analysis module also includes:

[0082] By calculating the risk index of the car driving at the i-th time point Compared with the preset risk index threshold Make a comparison;

[0083] like The system determines that the risk of the car driving at the i-th time point is low, and the vehicle is not prone to bumps and vibrations when driving, which does not affect the driver's control and judgment;

[0084] like The system determines that the car driving risk is high at the i-th time point. The vehicle is prone to strong bumps and vibrations when driving, which affects the driver's control and judgment. The system issues an early warning through the early warning module to remind the driver to reduce the speed.

[0085] Through the above technical solution, this embodiment calculates the risk index of the car driving at the i-th time point Compared with the preset risk index threshold By comparing, an accurate judgment can be made on the driving risk of the car at the i-th time point, and when it is judged that the driving risk of the car at the i-th time point is high, an early warning module is used to issue an early warning in time to remind the driver to reduce the speed. By setting it in this way, after combining vehicle information data and diversified data such as road cracks, the accuracy of the judgment results of the car driving risk can be improved, thereby ensuring the output of correct early warning information to the driver to avoid affecting the driver's judgment.

[0086] See also Figure 2 As shown, a road crack monitoring and early warning method based on machine vision, the method comprises:

[0087] S1: collecting road condition information in front of the vehicle at fixed time intervals through an image acquisition module arranged on the top of the vehicle;

[0088] S2: The image information collected by the image acquisition module is recognized by the image recognition module, and the road crack information in the image is extracted;

[0089] S3: collecting the vehicle status information and environmental data during driving through the data recording module and recording them;

[0090] S4: The road evaluation module calculates the abnormality index of the road ahead of the vehicle at different time points, and judges whether there is an abnormality on the current road in combination with the preset road abnormality index threshold, and the early warning module combines the analysis results to decide whether an early warning is needed;

[0091] S5: The risk analysis module combines the vehicle's own data with the evaluation results of the road evaluation module to analyze and judge the driving risk of the vehicle on the current road, and the early warning module combines the analysis results to decide whether an early warning is needed;

[0092] Through the above technical solution, this embodiment provides a road crack monitoring and early warning method based on machine vision, firstly, an image acquisition module arranged on the top of the vehicle is used to collect road condition information in front of the vehicle at fixed time intervals, then the image information collected by the image acquisition module is recognized by the image recognition module, and the road crack information in the image is extracted, and then the state information and environmental data of the vehicle during driving are collected by the data recording module, and the abnormality index of the road in front of the vehicle at different time points is calculated by the road evaluation module, and a judgment is made on whether there is an abnormality on the current road in combination with a preset road abnormality index threshold, and a decision is made on whether an early warning is needed by combining the analysis results with the early warning module, and finally the driving risk of the vehicle on the current road can be analyzed and judged by combining the vehicle's own data with the evaluation results of the road evaluation module, and a decision is made on whether an early warning is needed by combining the analysis results with the early warning module;

[0093] Through this method, the image acquisition module can first monitor road cracks, and by combining the road crack information with the vehicle's status information during driving for analysis, an accurate judgment can be made on the road condition information in front of the vehicle. Then, by combining the data with the vehicle's own data, the vehicle's driving risk can be further judged. The diversified data are linked together, thereby improving the accuracy of the judgment results, thereby ensuring that timely warnings can be issued to remind the driver in the event of road abnormalities or high driving risks, thereby increasing the driver's vigilance and preventing traffic accidents.

[0094] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. A road crack monitoring and early warning system based on machine vision, characterized in that: The system comprises: An image acquisition module, including a camera device disposed on the top of the vehicle, for acquiring road condition information in front of the vehicle at fixed time intervals; An image recognition module is used to recognize the image information collected by the image collection module and extract the road crack information in the image; The data recording module is used to collect and record the vehicle's status information and environmental data during driving; The road assessment module is used to combine the information obtained by the image recognition module with the vehicle's state information during driving, calculate the abnormality index of the road ahead of the vehicle at different time points, and judge whether there is an abnormality on the current road based on the preset road abnormality index threshold; The risk analysis module is used to analyze and judge the driving risk of the vehicle on the current road by combining the vehicle's own data with the evaluation results of the road evaluation module; The early warning module is used to combine the analysis results of the risk analysis module and the road assessment module to decide whether an early warning is needed.

2. The machine vision-based road crack monitoring and early warning system according to claim 1 is characterized in that: The road crack information identified by the image recognition module includes: road damage area, road damage length and maximum road damage depth; the data collected by the data recording module includes: vehicle speed, vibration amplitude, light intensity in the current shooting environment and dust content in the air during driving.

3. The machine vision-based road crack monitoring and early warning system according to claim 2 is characterized in that: The evaluation process of the road evaluation module includes: By formula Calculate the image loss coefficient δ at the i-th time point i ; Among them, i is a data collection at a fixed time interval, zd i is the vehicle vibration amplitude at the i-th time point, zd y is the preset vibration amplitude, sd i is the vehicle speed at the i-th time point, sd b sd i The standard value of μ is the error correction coefficient. According to the setting of empirical fitting, gx i is the light intensity at the ith time point, gx y is the preset light intensity, gx b For gx i The standard value, ca i is the dust content in the air at the i-th time point, ca b For ca i The standard value of , x1 and x2 are the first weight coefficients.

4. The machine vision-based road crack monitoring and early warning system according to claim 3 is characterized in that: The evaluation process of the road evaluation module also includes: By formula Calculate the abnormal index ρ of the road ahead of the vehicle at the i-th time point i ; Where a is any damaged area on the road ahead of the vehicle at the i-th time point, b is the total number of damaged areas on the road ahead of the vehicle at the i-th time point, and pos ai is the road damage area of ​​the ath damaged area in the road ahead of the vehicle at the i-th time point, z a is the weight coefficient of the a-th damaged area, which is set according to empirical fitting, γ a is the damage position influence coefficient of a damaged area, which is set according to empirical fitting, pos iz is the total area of ​​the road ahead of the vehicle at the i-th time point, sd ai is the road damage length of the ath damaged area in the road ahead of the vehicle at the i-th time point, sd y is the preset road damage length, cd i is the maximum depth of road damage in front of the vehicle at the i-th time point, cd y is the preset maximum depth of road damage, f t To adjust the coefficient lookup table function, according to the influence of the value range of the image loss coefficient at the i-th time point on the abnormality index of the road ahead of the vehicle, y1 and y2 are the second weight coefficients based on the test.

5. The machine vision-based road crack monitoring and early warning system according to claim 4 is characterized in that: The evaluation process of the road evaluation module also includes: By calculating the abnormal index ρ of the road ahead of the vehicle at the i-th time point i and the preset road anomaly coefficient threshold ρ 01 Make a comparison; If i ≤ρ 01 , the system determines that the abnormality index of the road ahead of the vehicle at the i-th time point is low, which means that the road condition ahead of the vehicle at the i-th time point is good, and no warning is issued; If i >ρ 01 The system determines that the abnormal index of the road ahead of the vehicle at the i-th time point is high, which means that the road condition ahead of the vehicle at the i-th time point is abnormal. The system issues an early warning through the early warning module and conducts risk analysis in combination with the vehicle's own data.

6. The machine vision-based road crack monitoring and early warning system according to claim 5, characterized in that: The analysis process of the risk analysis module includes: By formula Calculate the risk index of the car driving at the i-th time point Among them, dp is the chassis height of the car, dp b is the standard value of dp, xg is the suspension strength of the car, y is the preset suspension strength, lts is the usage time of the vehicle tire, lts y is the preset usage time of the tire, ω is the tire pressure influence coefficient, which is set based on empirical fitting, and jscl is the driver's driving experience.

7. The machine vision-based road crack monitoring and early warning system according to claim 6 is characterized in that: The analysis process of the risk analysis module also includes: By calculating the risk index of the car driving at the i-th time point Compared with the preset risk index threshold Make a comparison; like The system determines that the risk of the car driving at the i-th time point is low, and the vehicle is not prone to bumps and vibrations when driving, which does not affect the driver's control and judgment; like The system determines that the car driving risk is high at the i-th time point. The vehicle is prone to strong bumps and vibrations when driving, which affects the driver's control and judgment, and issues an early warning through the early warning module to remind the driver to slow down.

8. A road crack monitoring and early warning method based on machine vision, the method adopts a road crack monitoring and early warning system based on machine vision as claimed in claims 1-7, characterized in that: The method comprises: S1: collecting road condition information in front of the vehicle at fixed time intervals through an image acquisition module arranged on the top of the vehicle; S2: The image information collected by the image acquisition module is recognized by the image recognition module, and the road crack information in the image is extracted; S3: collecting the vehicle status information and environmental data during driving through the data recording module and recording them; S4: The road evaluation module calculates the abnormality index of the road ahead of the vehicle at different time points, and judges whether there is an abnormality on the current road in combination with the preset road abnormality index threshold, and the early warning module combines the analysis results to decide whether an early warning is needed; S5: The risk analysis module combines the vehicle's own data with the evaluation results of the road evaluation module to analyze and judge the driving risk of the vehicle on the current road, and the early warning module combines the analysis results to decide whether an early warning is needed.

Images