CN118196105B - Road quality detection method based on image recognition - Google Patents

Road quality detection method based on image recognition Download PDF

Info

Publication number
CN118196105B
CN118196105B CN202410619827.4A CN202410619827A CN118196105B CN 118196105 B CN118196105 B CN 118196105B CN 202410619827 A CN202410619827 A CN 202410619827A CN 118196105 B CN118196105 B CN 118196105B
Authority
CN
China
Prior art keywords
road
current detection
detection period
detection
period
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202410619827.4A
Other languages
Chinese (zh)
Other versions
CN118196105A (en
Inventor
罗小兵
谭松涛
王京涛
李磊
李涛
刘玉丰
郭文玉
袁悦
刘志博
黄梅彦
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Donggang Guangzeng Construction And Installation Co ltd
Original Assignee
Donggang Guangzeng Construction And Installation Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Donggang Guangzeng Construction And Installation Co ltd filed Critical Donggang Guangzeng Construction And Installation Co ltd
Priority to CN202410619827.4A priority Critical patent/CN118196105B/en
Publication of CN118196105A publication Critical patent/CN118196105A/en
Application granted granted Critical
Publication of CN118196105B publication Critical patent/CN118196105B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
    • Y02T10/00—Road transport of goods or passengers
    • Y02T10/10—Internal combustion engine [ICE] based vehicles
    • Y02T10/40—Engine management systems

Landscapes

  • Traffic Control Systems (AREA)

Abstract

The invention relates to the technical field of road quality detection, in particular to a road quality detection method based on image recognition; the detection road quality is detected and analyzed in four directions corresponding to the detected road in the current detection period, so that the road quality is improved conveniently based on the vehicle running state of the detected road, the state of the identification line in the road can be intuitively reflected, the safety problem of road running caused by unclear identification line in the road is avoided, and the accuracy and the practicability of the road quality detection result are greatly improved. And detecting the wheel trace data can effectively help the manager to know the apparent quality of the road in time, so that the scientificity and the effectiveness of the road quality detection result are greatly improved, the influence of foreign matters in the road on the running of the vehicle is avoided, and the foreign matters in the road can be timely reflected.

Description

Road quality detection method based on image recognition
Technical Field
The invention relates to the technical field of road quality detection, in particular to a road quality detection method based on image recognition.
Background
Roads are an important component of infrastructure, the quality of which is directly related to economic development. Good road quality not only promotes the flow of goods and personnel, but also promotes the development of business and travel. The road quality can directly influence the safety of the vehicle running, the excellent road quality not only can improve the comfort and the efficiency of the vehicle running and reduce the running resistance and the fuel consumption, but also reduces the possibility of accidents to a certain extent.
The current road quality detection is usually carried out manually in a set period, certain delay exists, the road quality state information cannot be obtained in real time, and further the follow-up adjustment of the road quality is not facilitated, so that the road quality cannot be effectively improved.
The detection and judgment of the mark lines and the wheel marks in the road are often omitted in the current road quality detection, so that the road quality detection result has one-sided property, and comprehensive analysis of the road quality is not facilitated, thereby influencing the quality inspection and maintenance of the road.
Disclosure of Invention
In view of the above-mentioned technical shortcomings, the present invention aims to provide a road quality detection method based on image recognition.
The aim of the invention can be achieved by the following technical scheme: a road quality detection method based on image recognition comprises the following steps:
step1, road vehicle data detection: detecting vehicle data corresponding to the detection road in the current detection period to obtain vehicle data corresponding to the detection road in the current detection period;
Preferably, the vehicle data corresponding to the detected road in the current detection period is detected by the following specific detection modes:
Acquiring vehicle images of the detection roads corresponding to all detection time points in the current detection period through the intelligent camera to obtain vehicle images of the detection roads corresponding to all detection time points in the current detection period;
identifying images of vehicles corresponding to the detection roads in the detection time points in the current detection time period from the vehicle images of the detection roads corresponding to the detection time points in the current detection time period, performing de-duplication treatment on the images to obtain images of vehicles corresponding to the detection roads in the current detection time period, taking the images as the traffic vehicles corresponding to the detection roads in the current detection time period, and further counting the number of the traffic vehicles corresponding to the detection roads in the current detection time period;
Extracting the sizes of the vehicles corresponding to the detected roads in the current detection period from the images of the vehicles corresponding to the detected roads in the current detection period based on the vehicles corresponding to the detected roads in the current detection period, and obtaining the sizes of the vehicles corresponding to the detected roads in the current detection period;
Detecting the weight of each passing vehicle corresponding to the detected road in the current detection period through a weight sensor to obtain the detected weight of each passing vehicle corresponding to the detected road in the current detection period;
Detecting the running speeds of the detected roads corresponding to the passing vehicles in the current detection period through interval speed measuring equipment to obtain the running speeds of the detected roads corresponding to the passing vehicles in the current detection period;
The number of vehicles corresponding to the detected road in the current detection period, the size of each vehicle, the detected weight of each vehicle and the running speed of each vehicle form the vehicle data corresponding to the detected road in the current detection period.
Step2, road vehicle data analysis: analyzing the vehicle influence index corresponding to the detection road in the current detection period based on the vehicle data corresponding to the detection road in the current detection period;
Preferably, the vehicle impact index corresponding to the detected road in the current detection period is analyzed based on the vehicle data corresponding to the detected road in the current detection period, and the specific analysis mode is as follows:
Extracting the number TN of passing vehicles corresponding to the detection road in the current detection period from the vehicle data corresponding to the detection road in the current detection period, extracting the size of each passing vehicle corresponding to the detection road in the current detection period, classifying the same size based on the size of each passing vehicle corresponding to the detection road in the current detection period to obtain each passing vehicle corresponding to each size of the detection road in the current detection period, and counting the number of each passing vehicle to obtain the number CN p of passing vehicles corresponding to each size of the detection road in the current detection period; p is the number of each size, p is a positive integer, p=1, 2,..w, w is the total number of size numbers;
Extracting the detection weight JG i of each passing vehicle corresponding to the detected road in the current detection period and the running speed XV i of each passing vehicle from the vehicle data corresponding to the detected road in the current detection period; i is the number of each passing vehicle, i is a positive integer, i=1, 2,..;
According to the formula Calculating a vehicle influence index CY corresponding to a detected road in the current detection period; a1, a2, a3, a4 and a5 are set influencing factors, the value ranges of a1, a2, a3, a4 and a5 are all more than 0 and less than 1, a p is a preset influence weight of a passing vehicle with the p-th size, and the value range of a p is more than 0 and less than 1.
Step3, road identification data detection: detecting the identification data corresponding to the detection road in the current detection period to obtain the identification data corresponding to the detection road in the current detection period;
Preferably, the identification data corresponding to the detected road in the current detection period is detected, so as to obtain the identification data corresponding to the detected road in the current detection period, and the specific detection mode is as follows:
Acquiring images of detection roads corresponding to all detection time points in the current detection period through a high-definition camera to obtain images of detection roads corresponding to all detection time points in the current detection period, extracting images of detection mark lines corresponding to the detection roads in all detection time points in the current detection period from the images, and performing de-duplication treatment on the images to obtain images of detection roads corresponding to all mark lines in the current detection period;
Obtaining ink breaking areas of the detection roads corresponding to the identification lines in the current detection period from the images of the detection roads corresponding to the identification lines in the current detection period, and obtaining the ink breaking areas of the detection roads corresponding to the identification lines in the current detection period;
Uniformly arranging detection points on the images of the detection roads corresponding to the identification lines in the current detection period, and collecting the chromaticity values of the detection points on the images of the detection roads corresponding to the identification lines in the current detection period by using a colorimeter to obtain the chromaticity values of the detection points of the detection roads corresponding to the identification lines in the current detection period;
and the ink breaking area of each detection road corresponding to each identification line in the current detection period and the chromaticity value of each detection point on each identification line form identification data corresponding to the detection road in the current detection period.
Step4, road identification data analysis: analyzing the identification evaluation index corresponding to the detected road in the current detection period based on the identification data corresponding to the detected road in the current detection period;
Preferably, the identification evaluation index corresponding to the detected road in the current detection period is analyzed based on the identification data corresponding to the detected road in the current detection period, and the specific analysis mode is as follows:
Extracting ink breaking areas DS j of the detection roads corresponding to the identification lines in the current detection period from the identification data corresponding to the detection roads in the current detection period, wherein j is the number of each identification line, j is a positive integer, j=1, 2, & gt, m and m are the total number of the identification line numbers; extracting chromaticity values SD j f of detection points on the corresponding identification lines of the detection road in the current detection period, wherein f is the number of each detection point, f is a positive integer, f=1, 2, & gt, g is the total number of the detection points;
Extracting a maximum chromaticity value SD j max and a minimum chromaticity value SD j min of the detection roads corresponding to the mark lines in the current detection period from the chromaticity values of the detection points of the detection roads corresponding to the mark lines in the current detection period;
calculating a chromaticity evaluation index SG j of each detection road corresponding to each identification line in the current detection period through a formula: E is a natural constant, SD j (f+1) is the chromaticity value of the f detection point on the j-th identification line corresponding to the detection road in the current detection period, SD 0 is the reference chromaticity value corresponding to the preset identification line, a6, a7 and a8 are set influence factors, and the value ranges of a6, a7 and a8 are all larger than 0 and smaller than 1;
According to the formula And calculating the corresponding identification evaluation indexes BZ of the detected road in the current detection period, wherein b1 and b2 are set influence factors, and the value ranges of b1 and b2 are larger than 0 and smaller than 1.
Step5, detecting apparent data of the road: detecting apparent data corresponding to the detection road in the current detection period to obtain the apparent data corresponding to the detection road in the current detection period;
Preferably, the apparent data corresponding to the detected road in the current detection period is detected, so as to obtain the apparent data corresponding to the detected road in the current detection period, and the specific detection mode is as follows:
Acquiring apparent images of detection roads corresponding to all detection time points in a current detection period through a high-definition camera to obtain apparent images of detection roads corresponding to all detection time points in the current detection period, identifying wheel trace images of the detection roads corresponding to all detection time points in the current detection period from the apparent images of the detection roads corresponding to all detection time points in the current detection period, and performing de-duplication processing on the wheel trace images to obtain wheel trace images corresponding to the detection roads in the current detection period;
Corresponding mark points are uniformly distributed on the image corresponding to each wheel trace of the detected road in the current detection period, and the width of each wheel trace of each mark point on each wheel trace of the detected road in the current detection period is obtained;
Acquiring the length of each wheel trace corresponding to the detected road in the current detection period from each wheel trace image corresponding to the detected road in the current detection period;
Identifying apparent cracks of the detection road corresponding to each detection time point in the current detection period from apparent images of the detection road corresponding to each detection time point in the current detection period, performing de-duplication treatment on the apparent cracks to obtain apparent cracks of the detection road in the current detection period, simultaneously obtaining crack areas of the detection road corresponding to the apparent cracks in the current detection period, and integrating the crack areas to obtain total crack areas of the detection road corresponding to the apparent cracks in the current detection period;
uniformly arranging test points on a detection road, importing a preset three-dimensional coordinate model, acquiring three-dimensional coordinates of each test point on the detection road based on the preset three-dimensional coordinate model, and extracting z-axis coordinate values of each test point on the detection road;
analyzing to obtain each concave point and each convex point corresponding to the detection road through a corresponding analysis mode, integrating adjacent concave points to obtain each concave position corresponding to the detection road, and integrating adjacent convex points to obtain each convex position corresponding to the detection road;
Collecting images of all concave parts and all convex parts of the detection road corresponding to all detection time points in the current detection period through the intelligent camera, and performing de-duplication treatment on the images to obtain images of all concave parts and images of all convex parts of the detection road in the current detection period;
Obtaining the concave area and the maximum concave depth of each concave corresponding to the detection road in the current detection period from the image of each concave corresponding to the detection road in the current detection period, and obtaining the total concave area, the number of concave positions and the maximum concave depth of each concave corresponding to the detection road in the current detection period through statistics;
The method comprises the steps of obtaining the protrusion area and the maximum protrusion height of each protrusion corresponding to the detection road in the current detection period from the image of each protrusion corresponding to the detection road in the current detection period, and obtaining the total protrusion area, the number of protrusions and the maximum protrusion height of each protrusion corresponding to the detection road in the current detection period through statistics;
The length of each wheel trace corresponding to the detected road in the current detection period, the width of each wheel trace of each mark point, the total crack area of the apparent crack, the total pit area, the number of pits, the maximum pit depth, the total bulge area, the number of bulges and the maximum bulge height form apparent data corresponding to the detected road in the current detection period.
Step6, analyzing apparent data of the road: analyzing the apparent evaluation index of the road corresponding to the detected road in the current detection period based on the apparent data corresponding to the detected road in the current detection period;
Preferably, the analysis is performed on the road appearance evaluation index corresponding to the detected road in the current detection period based on the appearance data corresponding to the detected road in the current detection period, and the specific analysis mode is as follows:
Extracting the length CL h of each wheel trace corresponding to the detected road in the current detection period and the wheel trace width CK h r of each marking point on each wheel trace from the apparent data corresponding to the detected road in the current detection period, wherein h is the number of each wheel trace, h is a positive integer, h=1, 2, y is the total number of the wheel trace numbers, r is the number of each marking point, r is a positive integer, r=1, 2, q and q are the total number of the marking point numbers;
According to the formula The wheel trace influence index DY corresponding to the detected road in the current detection period is calculated,For detecting the width of the wheel trace of the (r+1) th mark point of the h-th wheel trace of the road in the current detection period, b3, b4 and b5 are preset influence factors, and the value ranges of b3, b4 and b5 are all more than 0 and less than 1;
extracting the total crack area LS, the total pit area AS, the pit number AN, the maximum pit depth AH, the total bulge area TS, the number TP of bulges and the maximum bulge height TH of the apparent cracks of the detected road in the current detection period from the apparent data of the detected road in the current detection period, and obtaining the quality influence index GY of the apparent detection road in the current detection period through analysis;
According to the formula And calculating a road appearance evaluation index DG corresponding to the detected road in the current detection period, wherein c1 and c2 are set weight factors, and the value ranges of c1 and c2 are larger than 0 and smaller than 1.
Step7, road foreign matter identification analysis: identifying foreign matter data corresponding to the detected road in the current detection period to obtain the foreign matter data corresponding to the detected road in the current detection period, and analyzing a road foreign matter evaluation index corresponding to the detected road in the current detection period;
preferably, the foreign matter data corresponding to the detected road in the current detection period is identified in the following specific identification manner:
Acquiring images of detection roads corresponding to all detection time points in the current detection period by using a high-definition camera to obtain images of detection roads corresponding to all detection time points in the current detection period, and obtaining a foreign object image set corresponding to the detection roads in the current detection period by using corresponding analysis;
Performing de-duplication processing on the foreign object image set corresponding to the detection road in the current detection period to obtain each foreign object image corresponding to the detection road in the current detection period, and counting the number of foreign objects corresponding to the detection road and the volume of each foreign object in the current detection period;
The foreign matter data corresponding to the detected road in the current detection period is formed by the foreign matter quantity corresponding to the detected road in the current detection period and the volumes of the foreign matters.
Preferably, the road foreign matter evaluation index corresponding to the detected road in the current detection period is analyzed in the following specific analysis mode:
Extracting the number YN of the foreign matters corresponding to the detection road in the current detection period from the foreign matter data corresponding to the detection road in the current detection period, extracting the volumes of the foreign matters corresponding to the detection road in the current detection period, and summing the volumes to obtain the total volume YV of the foreign matters corresponding to the detection road in the current detection period;
According to the formula And calculating the road foreign matter evaluation indexes YY, c3 and c4 corresponding to the detected road in the current detection period to be set weight factors, wherein the value ranges of c3 and c4 are larger than 0 and smaller than 1.
Step8, road quality analysis: and analyzing the quality evaluation coefficient of the detected road corresponding to the current detection period based on the vehicle influence index, the identification evaluation index, the road appearance evaluation index and the road foreign matter evaluation index corresponding to the detected road in the current detection period to obtain the quality evaluation coefficient of the detected road corresponding to the current detection period.
Preferably, the quality evaluation coefficient of the detection road corresponding to the current detection period is analyzed, so as to obtain the quality evaluation coefficient of the detection road corresponding to the current detection period, and the specific analysis is as follows:
According to the formula And calculating the quality evaluation coefficients LZG, d1, d2, d3 and d4 of the detected road corresponding to the current detection period as set weight factors, wherein the value ranges of d1, d2, d3 and d4 are all larger than 0 and smaller than 1.
The invention has the beneficial effects that:
According to the invention, the vehicle data corresponding to the detection road in the current detection period is detected, and the vehicle influence index corresponding to the detection road in the current detection period is analyzed based on the detection data, so that the running state of the vehicle in the detection road can be reflected in real time, and the road quality can be improved based on the running state of the vehicle in the detection road conveniently, such as road widening, road extending and the like.
According to the invention, the identification data corresponding to the detected road in the current detection period is detected, and the identification evaluation index corresponding to the detected road in the current detection period is obtained based on the analysis, so that the state of the identification line in the road can be intuitively reflected, the safety problem of road running caused by unclear identification line in the road is avoided, and the accuracy and the practicability of the road quality detection result are greatly improved to a certain extent.
According to the invention, the apparent data corresponding to the detected road in the current detection period are detected, and the apparent evaluation index of the road corresponding to the detected road in the current detection period is obtained based on the analysis, so that the abrasion of the road can be accelerated to a certain extent by the wheel trace, and the detection of the wheel trace data can effectively help management staff to know the apparent quality problem of the road in time, so that the corresponding adjustment and improvement are carried out, and the scientificity and the effectiveness of the road quality detection result are greatly improved.
According to the invention, the foreign matter data corresponding to the detected road in the current detection period is identified and analyzed, so that the influence of the foreign matters in the road on the running of the vehicle is avoided to a certain extent, the foreign matter condition in the road can be reflected timely, and further, the management personnel can clean and clean timely.
Drawings
The invention is further described below with reference to the accompanying drawings.
FIG. 1 is a flow chart of the method steps of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1, the invention provides a road quality detection method based on image recognition, which comprises the following steps:
Step1, road vehicle data detection: detecting the vehicle data corresponding to the detection road in the current detection period to obtain the vehicle data corresponding to the detection road in the current detection period, wherein the specific detection mode is as follows:
Acquiring vehicle images of the detection roads corresponding to all detection time points in the current detection period through the intelligent camera to obtain vehicle images of the detection roads corresponding to all detection time points in the current detection period;
identifying images of vehicles corresponding to the detection roads in the detection time points in the current detection time period from the vehicle images of the detection roads corresponding to the detection time points in the current detection time period, performing de-duplication treatment on the images to obtain images of vehicles corresponding to the detection roads in the current detection time period, taking the images as the traffic vehicles corresponding to the detection roads in the current detection time period, and further counting the number of the traffic vehicles corresponding to the detection roads in the current detection time period;
Extracting the sizes of the vehicles corresponding to the detected roads in the current detection period from the images of the vehicles corresponding to the detected roads in the current detection period based on the vehicles corresponding to the detected roads in the current detection period, and obtaining the sizes of the vehicles corresponding to the detected roads in the current detection period;
The dimensions of the passing vehicle are specifically as follows: large, medium, small and medium. Wherein the size of the passing vehicle is judged based on the height and width of the passing vehicle, for example: the large-sized trucks are large in size, and the cars are small in size.
Detecting the weight of each passing vehicle corresponding to the detected road in the current detection period through a weight sensor to obtain the detected weight of each passing vehicle corresponding to the detected road in the current detection period;
In a specific embodiment, the weight of each passing vehicle corresponding to the road is detected in the current detection period, and the specific detection mode is as follows: weight sensors are arranged at the starting point and the end point of the detected road, the weight of each passing vehicle is detected through the weight sensor for detecting the starting point and the weight sensor for detecting the end point in the road, and the average value is taken as the detected weight of each passing vehicle.
Detecting the running speeds of the detected roads corresponding to the passing vehicles in the current detection period through interval speed measuring equipment to obtain the running speeds of the detected roads corresponding to the passing vehicles in the current detection period;
The number of vehicles corresponding to the detected road in the current detection period, the size of each vehicle, the detected weight of each vehicle and the running speed of each vehicle form the vehicle data corresponding to the detected road in the current detection period.
Step2, road vehicle data analysis: the vehicle influence indexes corresponding to the detection roads in the current detection period are analyzed based on the vehicle data corresponding to the detection roads in the current detection period, and the specific analysis mode is as follows:
Extracting the number TN of passing vehicles corresponding to the detection road in the current detection period from the vehicle data corresponding to the detection road in the current detection period, extracting the size of each passing vehicle corresponding to the detection road in the current detection period, classifying the same size based on the size of each passing vehicle corresponding to the detection road in the current detection period to obtain each passing vehicle corresponding to each size of the detection road in the current detection period, and counting the number of each passing vehicle to obtain the number CN p of passing vehicles corresponding to each size of the detection road in the current detection period; p is the number of each size, p is a positive integer, p=1, 2,..w, w is the total number of size numbers;
Extracting the detection weight JG i of each passing vehicle corresponding to the detected road in the current detection period and the running speed XV i of each passing vehicle from the vehicle data corresponding to the detected road in the current detection period; i is the number of each passing vehicle, i is a positive integer, i=1, 2,..;
According to the formula Calculating a vehicle influence index CY corresponding to a detected road in the current detection period; a1, a2, a3, a4 and a5 are set influencing factors, the value ranges of a1, a2, a3, a4 and a5 are all more than 0 and less than 1, a p is a preset influence weight of a passing vehicle with the p-th size, and the value range of a p is more than 0 and less than 1.
It should be noted that, the number of vehicles corresponding to the detected road in the current detection period, the detected weight of each vehicle, and the running speed of each vehicle are normalized, and then the values are obtained, and then the values are substituted into the calculation formula of the vehicle impact index corresponding to the detected road in the current detection period for calculation.
Step3, road identification data detection: detecting the identification data corresponding to the detection road in the current detection period to obtain the identification data corresponding to the detection road in the current detection period, wherein the specific detection mode is as follows:
Acquiring images of detection roads corresponding to all detection time points in the current detection period through a high-definition camera to obtain images of detection roads corresponding to all detection time points in the current detection period, extracting images of detection mark lines corresponding to the detection roads in all detection time points in the current detection period from the images, and performing de-duplication treatment on the images to obtain images of detection roads corresponding to all mark lines in the current detection period;
Obtaining ink breaking areas of the detection roads corresponding to the identification lines in the current detection period from the images of the detection roads corresponding to the identification lines in the current detection period, and obtaining the ink breaking areas of the detection roads corresponding to the identification lines in the current detection period;
Uniformly arranging detection points on the images of the detection roads corresponding to the identification lines in the current detection period, and collecting the chromaticity values of the detection points on the images of the detection roads corresponding to the identification lines in the current detection period by using a colorimeter to obtain the chromaticity values of the detection points of the detection roads corresponding to the identification lines in the current detection period;
and the ink breaking area of each detection road corresponding to each identification line in the current detection period and the chromaticity value of each detection point on each identification line form identification data corresponding to the detection road in the current detection period.
Step4, road identification data analysis: analyzing the identification evaluation index corresponding to the detected road in the current detection period based on the identification data corresponding to the detected road in the current detection period, wherein the specific analysis mode is as follows:
Extracting ink breaking areas DS j of the detection roads corresponding to the identification lines in the current detection period from the identification data corresponding to the detection roads in the current detection period, wherein j is the number of each identification line, j is a positive integer, j=1, 2, & gt, m and m are the total number of the identification line numbers; extracting chromaticity values SD j f of detection points on the corresponding identification lines of the detection road in the current detection period, wherein f is the number of each detection point, f is a positive integer, f=1, 2, & gt, g is the total number of the detection points;
Extracting a maximum chromaticity value SD j max and a minimum chromaticity value SD j min of the detection roads corresponding to the mark lines in the current detection period from the chromaticity values of the detection points of the detection roads corresponding to the mark lines in the current detection period;
calculating a chromaticity evaluation index SG j of each detection road corresponding to each identification line in the current detection period through a formula: E is a natural constant, SD j (f+1) is the chromaticity value of the f detection point on the j-th identification line corresponding to the detection road in the current detection period, SD 0 is the reference chromaticity value corresponding to the preset identification line, a6, a7 and a8 are set influence factors, and the value ranges of a6, a7 and a8 are all larger than 0 and smaller than 1;
According to the formula And calculating the corresponding identification evaluation indexes BZ of the detected road in the current detection period, wherein b1 and b2 are set influence factors, and the value ranges of b1 and b2 are larger than 0 and smaller than 1.
Step5, detecting apparent data of the road: detecting apparent data corresponding to the detected road in the current detection period to obtain the apparent data corresponding to the detected road in the current detection period, wherein the specific detection mode is as follows:
Acquiring apparent images of detection roads corresponding to all detection time points in a current detection period through a high-definition camera to obtain apparent images of detection roads corresponding to all detection time points in the current detection period, identifying wheel trace images of the detection roads corresponding to all detection time points in the current detection period from the apparent images of the detection roads corresponding to all detection time points in the current detection period, and performing de-duplication processing on the wheel trace images to obtain wheel trace images corresponding to the detection roads in the current detection period;
Corresponding mark points are uniformly distributed on the image corresponding to each wheel trace of the detected road in the current detection period, and the width of each wheel trace of each mark point on each wheel trace of the detected road in the current detection period is obtained;
it should be noted that, the method for obtaining the width of the wheel trace corresponding to each mark point on each wheel trace in the road is specifically:
And uniformly distributing marking points at the left end or the right end of the image corresponding to each wheel trace of the detected road in the current detection period to obtain each marking point on the image corresponding to each wheel trace of the detected road in the current detection period, and acquiring the width of each marking point corresponding to each wheel trace of the detected road in the current detection period based on each marking point on the image corresponding to each wheel trace of the detected road in the current detection period.
Acquiring the length of each wheel trace corresponding to the detected road in the current detection period from each wheel trace image corresponding to the detected road in the current detection period;
It should be noted that, the length of each wheel trace corresponding to the detected road in the current detection period is specifically obtained by: and arranging a plurality of detection points at the bottom end or the top end of each wheel trace image corresponding to the detection road in the current detection period to obtain each detection point of each wheel trace image corresponding to the detection road in the current detection period, acquiring the length of each detection point of each wheel trace image corresponding to the detection road in the current detection period based on each detection point of each wheel trace image corresponding to the detection road in the current detection period, sequentially arranging the lengths of each detection point of each wheel trace image corresponding to the detection road in the current detection period from large to small, removing the maximum length, the minimum length and the middle length, and taking the average value as the length of each wheel trace corresponding to the detection road in the current detection period.
The middle length is a value arranged in the middle in the arrangement sequence of the detection points of the detection road corresponding to the trace images of each wheel in the current detection period, and it is noted that if the number of the values arranged in the middle is two, the two values arranged in the middle are both called as the middle length.
Identifying apparent cracks of the detection road corresponding to each detection time point in the current detection period from apparent images of the detection road corresponding to each detection time point in the current detection period, performing de-duplication treatment on the apparent cracks to obtain apparent cracks of the detection road in the current detection period, simultaneously obtaining crack areas of the detection road corresponding to the apparent cracks in the current detection period, and integrating the crack areas to obtain total crack areas of the detection road corresponding to the apparent cracks in the current detection period;
uniformly arranging test points on a detection road, importing a preset three-dimensional coordinate model, acquiring three-dimensional coordinates of each test point on the detection road based on the preset three-dimensional coordinate model, and extracting z-axis coordinate values of each test point on the detection road;
analyzing to obtain each concave point and each convex point corresponding to the detection road through a corresponding analysis mode, integrating adjacent concave points to obtain each concave position corresponding to the detection road, and integrating adjacent convex points to obtain each convex position corresponding to the detection road;
Comparing the z-axis coordinate value of each test point on the detection road with a preset reference z-axis coordinate value, executing 001 if the z-axis coordinate value of a certain test point on the detection road is larger than the preset reference z-axis coordinate value, and executing 002 if the z-axis coordinate value of a certain test point on the detection road is smaller than the preset reference z-axis coordinate value;
001: the z-axis coordinate value of the test point is differenced with a preset reference z-axis coordinate value to obtain a z-axis coordinate difference value of the test point, the z-axis coordinate difference value of the test point is compared with a preset primary coordinate difference value, and if the z-axis coordinate difference value of the test point is larger than the preset primary coordinate difference value, the test point is marked as a bulge point;
002: performing difference between the z-axis coordinate value of the test point and a preset reference z-axis coordinate value to obtain a z-axis coordinate difference value of the test point, comparing the z-axis coordinate difference value of the test point with a preset secondary coordinate difference value, and if the z-axis coordinate difference value of the test point is smaller than the preset secondary coordinate difference value, marking the test point as a concave point; and thus, each concave point and each convex point corresponding to the detected road are obtained through statistics.
Integrating adjacent concave points to obtain corresponding concave positions of the detection road, and integrating adjacent convex points to obtain corresponding convex positions of the detection road;
Collecting images of all concave parts and all convex parts of the detection road corresponding to all detection time points in the current detection period through the intelligent camera, and performing de-duplication treatment on the images to obtain images of all concave parts and images of all convex parts of the detection road in the current detection period;
Obtaining the concave area and the maximum concave depth of each concave corresponding to the detection road in the current detection period from the image of each concave corresponding to the detection road in the current detection period, and obtaining the total concave area, the number of concave positions and the maximum concave depth of each concave corresponding to the detection road in the current detection period through statistics;
The method comprises the steps of obtaining the protrusion area and the maximum protrusion height of each protrusion corresponding to the detection road in the current detection period from the image of each protrusion corresponding to the detection road in the current detection period, and obtaining the total protrusion area, the number of protrusions and the maximum protrusion height of each protrusion corresponding to the detection road in the current detection period through statistics;
The length of each wheel trace corresponding to the detected road in the current detection period, the width of each wheel trace of each mark point, the total crack area of the apparent crack, the total pit area, the number of pits, the maximum pit depth, the total bulge area, the number of bulges and the maximum bulge height form apparent data corresponding to the detected road in the current detection period.
Step6, analyzing apparent data of the road: analyzing the road appearance evaluation index corresponding to the detected road in the current detection period based on the appearance data corresponding to the detected road in the current detection period, wherein the specific analysis mode is as follows:
Extracting the length CL h of each wheel trace corresponding to the detected road in the current detection period and the wheel trace width CK h r of each marking point on each wheel trace from the apparent data corresponding to the detected road in the current detection period, wherein h is the number of each wheel trace, h is a positive integer, h=1, 2, y is the total number of the wheel trace numbers, r is the number of each marking point, r is a positive integer, r=1, 2, q and q are the total number of the marking point numbers;
According to the formula The wheel trace influence index DY corresponding to the detected road in the current detection period is calculated,For detecting the width of the wheel trace of the (r+1) th mark point of the h-th wheel trace of the road in the current detection period, b3, b4 and b5 are preset influence factors, and the value ranges of b3, b4 and b5 are all more than 0 and less than 1;
extracting the total crack area LS, the total pit area AS, the pit number AN, the maximum pit depth AH, the total bulge area TS, the number TP of bulges and the maximum bulge height TH of the apparent cracks of the detected road in the current detection period from the apparent data of the detected road in the current detection period, and obtaining the quality influence index GY of the apparent detection road in the current detection period through analysis;
According to the formula And calculating the quality influence indexes GY, b6, b7, b8, b9, ba, bb and bc of the corresponding appearance of the detected road in the current detection period to be set weight factors, wherein the value ranges of b6, b7, b8, b9, ba, bb and bc are all larger than 0 and smaller than 1.
According to the formulaAnd calculating a road appearance evaluation index DG corresponding to the detected road in the current detection period, wherein c1 and c2 are set weight factors, and the value ranges of c1 and c2 are larger than 0 and smaller than 1.
Step7, road foreign matter identification analysis: identifying foreign matter data corresponding to the detected road in the current detection period to obtain the foreign matter data corresponding to the detected road in the current detection period, and analyzing a road foreign matter evaluation index corresponding to the detected road in the current detection period, wherein the specific detection and analysis are as follows:
acquiring images of detection roads corresponding to all detection time points in a current detection period through a high-definition camera to obtain images of detection roads corresponding to all detection time points in the current detection period, matching the images of the detection roads corresponding to all detection time points in the current detection period with a preset normal reference image set, and if the images of the detection roads corresponding to a certain detection time point are not successfully matched with the preset normal reference image set, marking the images of the detection roads corresponding to the detection points as foreign matter images, so that the images of the foreign matter images corresponding to the detection roads in the current detection period are obtained and serve as the foreign matter image set corresponding to the detection roads in the current detection period.
Note that the normal reference image includes: the vehicle is traveling in a road, the vehicle is parked, etc. Foreign matter includes, but is not limited to: stone, glass fragments, branches.
Performing de-duplication processing on the foreign object image set corresponding to the detection road in the current detection period to obtain each foreign object image corresponding to the detection road in the current detection period, and counting the number of foreign objects corresponding to the detection road and the volume of each foreign object in the current detection period;
The foreign matter data corresponding to the detected road in the current detection period is formed by the foreign matter quantity corresponding to the detected road in the current detection period and the volumes of the foreign matters.
Extracting the number YN of the foreign matters corresponding to the detection road in the current detection period from the foreign matter data corresponding to the detection road in the current detection period, extracting the volumes of the foreign matters corresponding to the detection road in the current detection period, and summing the volumes to obtain the total volume YV of the foreign matters corresponding to the detection road in the current detection period;
According to the formula And calculating the road foreign matter evaluation indexes YY, c3 and c4 corresponding to the detected road in the current detection period to be set weight factors, wherein the value ranges of c3 and c4 are larger than 0 and smaller than 1.
Step8, road quality analysis: based on the vehicle influence index, the identification evaluation index, the road appearance evaluation index and the road foreign matter evaluation index corresponding to the detected road in the current detection period, analyzing the quality evaluation coefficient of the detected road corresponding to the current detection period to obtain the quality evaluation coefficient of the detected road corresponding to the current detection period, wherein the specific calculation formula is as follows: LZG is a quality evaluation coefficient of a detection road corresponding to the current detection period, d1, d2, d3 and d4 are set weight factors, and the value ranges of d1, d2, d3 and d4 are all larger than 0 and smaller than 1.
The foregoing is merely illustrative of the structures of this invention and various modifications, additions and substitutions for those skilled in the art of describing particular embodiments without departing from the structures of the invention or exceeding the scope of the invention as defined in the accompanying drawings are intended to be included within the scope of the invention.

Claims (7)

1. The road quality detection method based on image recognition is characterized by comprising the following steps of:
step1, road vehicle data detection: detecting vehicle data corresponding to the detection road in the current detection period to obtain vehicle data corresponding to the detection road in the current detection period;
Step2, road vehicle data analysis: analyzing the vehicle influence indexes corresponding to the detected roads in the current detection period based on the vehicle data corresponding to the detected roads in the current detection period:
Extracting the number TN of passing vehicles corresponding to the detection road in the current detection period from the vehicle data corresponding to the detection road in the current detection period, extracting the size of each passing vehicle corresponding to the detection road in the current detection period, classifying the same size based on the size of each passing vehicle corresponding to the detection road in the current detection period to obtain each passing vehicle corresponding to each size of the detection road in the current detection period, and counting the number of each passing vehicle to obtain the number CN p of passing vehicles corresponding to each size of the detection road in the current detection period; p is the number of each size, p is a positive integer, p=1, 2,..w, w is the total number of size numbers;
Extracting the detection weight JG i of each passing vehicle corresponding to the detected road in the current detection period and the running speed XV i of each passing vehicle from the vehicle data corresponding to the detected road in the current detection period; i is the number of each passing vehicle, i is a positive integer, i=1, 2,..;
According to the formula Calculating a vehicle influence index CY corresponding to a detected road in the current detection period; a1, a2, a3, a4 and a5 are set influence factors, the value ranges of a1, a2, a3, a4 and a5 are all more than 0 and less than 1, a p is a preset influence weight of a passing vehicle with the p-th size, and the value range of a p is more than 0 and less than 1;
step3, road identification data detection: detecting the identification data corresponding to the detection road in the current detection period to obtain the identification data corresponding to the detection road in the current detection period;
Step4, road identification data analysis: analyzing the identification evaluation index corresponding to the detected road in the current detection period based on the identification data corresponding to the detected road in the current detection period:
Extracting ink breaking areas DS j of the detection roads corresponding to the identification lines in the current detection period from the identification data corresponding to the detection roads in the current detection period, wherein j is the number of each identification line, j is a positive integer, j=1, 2, & gt, m and m are the total number of the identification line numbers; extracting chromaticity values SD j f of detection points on the corresponding identification lines of the detection road in the current detection period, wherein f is the number of each detection point, f is a positive integer, f=1, 2, & gt, g is the total number of the detection points;
Extracting a maximum chromaticity value SD j max and a minimum chromaticity value SD j min of the detection roads corresponding to the mark lines in the current detection period from the chromaticity values of the detection points of the detection roads corresponding to the mark lines in the current detection period;
Obtaining a chromaticity evaluation index SG j of each identification line corresponding to the detection road in the current detection period through analysis;
According to the formula Calculating an identification evaluation index BZ corresponding to a detection road in a current detection period, wherein b1 and b2 are set influence factors, and the value ranges of b1 and b2 are both larger than 0 and smaller than 1;
Step5, detecting apparent data of the road: detecting apparent data corresponding to the detection road in the current detection period to obtain the apparent data corresponding to the detection road in the current detection period;
Step6, analyzing apparent data of the road: analyzing the road appearance evaluation index corresponding to the detected road in the current detection period based on the appearance data corresponding to the detected road in the current detection period:
Extracting the length CL h of each wheel trace corresponding to the detected road in the current detection period and the wheel trace width CK h r of each marking point on each wheel trace from the apparent data corresponding to the detected road in the current detection period, wherein h is the number of each wheel trace, h is a positive integer, h=1, 2, y is the total number of the wheel trace numbers, r is the number of each marking point, r is a positive integer, r=1, 2, q and q are the total number of the marking point numbers;
According to the formula The wheel trace influence index DY corresponding to the detected road in the current detection period is calculated,For detecting the width of the wheel trace of the (r+1) th mark point of the h-th wheel trace of the road in the current detection period, b3, b4 and b5 are preset influence factors, and the value ranges of b3, b4 and b5 are all more than 0 and less than 1;
Extracting the total crack area, the total pit area, the number of pits, the maximum pit depth, the total protrusion area, the number of protrusions and the maximum protrusion height of the apparent cracks corresponding to the detected road in the current detection period from the apparent data corresponding to the detected road in the current detection period, and obtaining a quality influence index GY of the apparent corresponding to the detected road in the current detection period through analysis;
According to the formula Calculating a road appearance evaluation index DG corresponding to a detected road in a current detection period, wherein c1 and c2 are set weight factors, and the value ranges of c1 and c2 are larger than 0 and smaller than 1;
step7, road foreign matter identification analysis: identifying foreign matter data corresponding to the detected road in the current detection period to obtain the foreign matter data corresponding to the detected road in the current detection period, and analyzing a road foreign matter evaluation index corresponding to the detected road in the current detection period;
Step8, road quality analysis: and analyzing the quality evaluation coefficient of the detected road corresponding to the current detection period based on the vehicle influence index, the identification evaluation index, the road appearance evaluation index and the road foreign matter evaluation index corresponding to the detected road in the current detection period to obtain the quality evaluation coefficient of the detected road corresponding to the current detection period.
2. The road quality detection method based on image recognition according to claim 1, wherein the detecting the vehicle data corresponding to the detected road in the current detection period comprises the following specific detection modes:
Acquiring vehicle images of the detection roads corresponding to all detection time points in the current detection period through the intelligent camera to obtain vehicle images of the detection roads corresponding to all detection time points in the current detection period;
identifying images of vehicles corresponding to the detection roads in the detection time points in the current detection time period from the vehicle images of the detection roads corresponding to the detection time points in the current detection time period, performing de-duplication treatment on the images to obtain images of vehicles corresponding to the detection roads in the current detection time period, taking the images as the traffic vehicles corresponding to the detection roads in the current detection time period, and further counting the number of the traffic vehicles corresponding to the detection roads in the current detection time period;
Extracting the sizes of the vehicles corresponding to the detected roads in the current detection period from the images of the vehicles corresponding to the detected roads in the current detection period based on the vehicles corresponding to the detected roads in the current detection period, and obtaining the sizes of the vehicles corresponding to the detected roads in the current detection period;
Detecting the weight of each passing vehicle corresponding to the detected road in the current detection period through a weight sensor to obtain the detected weight of each passing vehicle corresponding to the detected road in the current detection period;
Detecting the running speeds of the detected roads corresponding to the passing vehicles in the current detection period through interval speed measuring equipment to obtain the running speeds of the detected roads corresponding to the passing vehicles in the current detection period;
The number of vehicles corresponding to the detected road in the current detection period, the size of each vehicle, the detected weight of each vehicle and the running speed of each vehicle form the vehicle data corresponding to the detected road in the current detection period.
3. The method for detecting road quality based on image recognition according to claim 1, wherein the detecting the identification data corresponding to the detected road in the current detection period is performed to obtain the identification data corresponding to the detected road in the current detection period, and the specific detecting mode is as follows:
Acquiring images of detection roads corresponding to all detection time points in the current detection period through a high-definition camera to obtain images of detection roads corresponding to all detection time points in the current detection period, extracting images of detection mark lines corresponding to the detection roads in all detection time points in the current detection period from the images, and performing de-duplication treatment on the images to obtain images of detection roads corresponding to all mark lines in the current detection period;
Obtaining ink breaking areas of the detection roads corresponding to the identification lines in the current detection period from the images of the detection roads corresponding to the identification lines in the current detection period, and obtaining the ink breaking areas of the detection roads corresponding to the identification lines in the current detection period;
Uniformly arranging detection points on the images of the detection roads corresponding to the identification lines in the current detection period, and collecting the chromaticity values of the detection points on the images of the detection roads corresponding to the identification lines in the current detection period by using a colorimeter to obtain the chromaticity values of the detection points of the detection roads corresponding to the identification lines in the current detection period;
and the ink breaking area of each detection road corresponding to each identification line in the current detection period and the chromaticity value of each detection point on each identification line form identification data corresponding to the detection road in the current detection period.
4. The method for detecting road quality based on image recognition according to claim 1, wherein the detecting the apparent data corresponding to the detected road in the current detecting period is performed to obtain the apparent data corresponding to the detected road in the current detecting period, and the specific detecting mode is as follows:
Acquiring apparent images of detection roads corresponding to all detection time points in a current detection period through a high-definition camera to obtain apparent images of detection roads corresponding to all detection time points in the current detection period, identifying wheel trace images of the detection roads corresponding to all detection time points in the current detection period from the apparent images of the detection roads corresponding to all detection time points in the current detection period, and performing de-duplication processing on the wheel trace images to obtain wheel trace images corresponding to the detection roads in the current detection period;
Corresponding mark points are uniformly distributed on the image corresponding to each wheel trace of the detected road in the current detection period, and the width of each wheel trace of each mark point on each wheel trace of the detected road in the current detection period is obtained;
Acquiring the length of each wheel trace corresponding to the detected road in the current detection period from each wheel trace image corresponding to the detected road in the current detection period;
Identifying apparent cracks of the detection road corresponding to each detection time point in the current detection time period from apparent images of the detection road corresponding to each detection time point in the current detection time period, and obtaining the total crack area of the apparent cracks of the detection road corresponding to the current detection time period through analysis;
uniformly arranging test points on a detection road, importing a preset three-dimensional coordinate model, acquiring three-dimensional coordinates of each test point on the detection road based on the preset three-dimensional coordinate model, and extracting z-axis coordinate values of each test point on the detection road;
analyzing to obtain each concave point and each convex point corresponding to the detection road through a corresponding analysis mode, integrating adjacent concave points to obtain each concave position corresponding to the detection road, and integrating adjacent convex points to obtain each convex position corresponding to the detection road;
Collecting images of all concave parts and all convex parts of the detection road corresponding to all detection time points in the current detection period through the intelligent camera, and performing de-duplication treatment on the images to obtain images of all concave parts and images of all convex parts of the detection road in the current detection period;
Obtaining the concave area and the maximum concave depth of each concave corresponding to the detection road in the current detection period from the image of each concave corresponding to the detection road in the current detection period, and obtaining the total concave area, the number of concave positions and the maximum concave depth of each concave corresponding to the detection road in the current detection period through statistics;
The method comprises the steps of obtaining the protrusion area and the maximum protrusion height of each protrusion corresponding to the detection road in the current detection period from the image of each protrusion corresponding to the detection road in the current detection period, and obtaining the total protrusion area, the number of protrusions and the maximum protrusion height of each protrusion corresponding to the detection road in the current detection period through statistics;
The length of each wheel trace corresponding to the detected road in the current detection period, the width of each wheel trace of each mark point, the total crack area of the apparent crack, the total pit area, the number of pits, the maximum pit depth, the total bulge area, the number of bulges and the maximum bulge height form apparent data corresponding to the detected road in the current detection period.
5. The method for detecting road quality based on image recognition according to claim 1, wherein the specific recognition mode for recognizing the foreign matter data corresponding to the detected road in the current detection period is as follows:
Acquiring images of detection roads corresponding to all detection time points in the current detection period by using a high-definition camera to obtain images of detection roads corresponding to all detection time points in the current detection period, and obtaining a foreign object image set corresponding to the detection roads in the current detection period by using corresponding analysis;
Performing de-duplication processing on the foreign object image set corresponding to the detection road in the current detection period to obtain each foreign object image corresponding to the detection road in the current detection period, and counting the number of foreign objects corresponding to the detection road and the volume of each foreign object in the current detection period;
The foreign matter data corresponding to the detected road in the current detection period is formed by the foreign matter quantity corresponding to the detected road in the current detection period and the volumes of the foreign matters.
6. The road quality detection method based on image recognition according to claim 1, wherein the analysis is performed on the road foreign matter evaluation index corresponding to the detected road in the current detection period by the following specific analysis method:
Extracting the number YN of the foreign matters corresponding to the detection road in the current detection period from the foreign matter data corresponding to the detection road in the current detection period, extracting the volumes of the foreign matters corresponding to the detection road in the current detection period, and summing the volumes to obtain the total volume YV of the foreign matters corresponding to the detection road in the current detection period;
According to the formula And calculating the road foreign matter evaluation indexes YY, c3 and c4 corresponding to the detected road in the current detection period to be set weight factors, wherein the value ranges of c3 and c4 are larger than 0 and smaller than 1.
7. The method for detecting road quality based on image recognition according to claim 6, wherein the analyzing the quality evaluation coefficient of the detected road corresponding to the current detection period to obtain the quality evaluation coefficient of the detected road corresponding to the current detection period specifically includes:
According to the formula And calculating the quality evaluation coefficients LZG, d1, d2, d3 and d4 of the detected road corresponding to the current detection period as set weight factors, wherein the value ranges of d1, d2, d3 and d4 are all larger than 0 and smaller than 1.
CN202410619827.4A 2024-05-20 2024-05-20 Road quality detection method based on image recognition Active CN118196105B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202410619827.4A CN118196105B (en) 2024-05-20 2024-05-20 Road quality detection method based on image recognition

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202410619827.4A CN118196105B (en) 2024-05-20 2024-05-20 Road quality detection method based on image recognition

Publications (2)

Publication Number Publication Date
CN118196105A CN118196105A (en) 2024-06-14
CN118196105B true CN118196105B (en) 2024-07-30

Family

ID=91404681

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202410619827.4A Active CN118196105B (en) 2024-05-20 2024-05-20 Road quality detection method based on image recognition

Country Status (1)

Country Link
CN (1) CN118196105B (en)

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114758322A (en) * 2022-05-13 2022-07-15 安徽省路通公路工程检测有限公司 Road quality detection system based on machine identification
CN115423868A (en) * 2022-08-23 2022-12-02 深圳泰豪信息技术有限公司 Spatial positioning linkage calibration method of video monitoring camera

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FR3135945A1 (en) * 2022-05-25 2023-12-01 Renault S.A.S. Method for detecting a type of road traveled by an equipped vehicle
CN116343441B (en) * 2023-03-29 2023-11-24 安徽光谷智能科技股份有限公司 Expressway safety monitoring system based on multidimensional real-time monitoring
CN116153078B (en) * 2023-04-14 2023-08-01 健鼎（无锡）电子有限公司 Road safety assessment method and device based on millimeter wave radar and storage medium
CN117173899B (en) * 2023-11-03 2024-01-30 武汉网信安全技术股份有限公司 Smart city traffic travel quality evaluation method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114758322A (en) * 2022-05-13 2022-07-15 安徽省路通公路工程检测有限公司 Road quality detection system based on machine identification
CN115423868A (en) * 2022-08-23 2022-12-02 深圳泰豪信息技术有限公司 Spatial positioning linkage calibration method of video monitoring camera

Also Published As

Publication number Publication date
CN118196105A (en) 2024-06-14

Similar Documents

Publication Publication Date Title
CN100403332C (en) Vehicle lane Robust identifying method for lane deviation warning
CN103324913B (en) A kind of pedestrian event detection method of Shape-based interpolation characteristic sum trajectory analysis
CN103034836B (en) Road sign detection method and road sign checkout equipment
CN103279996B (en) Information of vehicles in a kind of multilane situation detects and recognition system
CN104966049B (en) Lorry detection method based on image
CN102867417B (en) Taxi anti-forgery system and taxi anti-forgery method
CN115717894B (en) Vehicle high-precision positioning method based on GPS and common navigation map
CN101944176A (en) Exist the more excellent clear path of means of transportation sign to detect
CN105005771A (en) Method for detecting full line of lane based on optical flow point locus statistics
CN104376297A (en) Detection method and device for linear indication signs on road
CN109910881B (en) Turning road vision blind area information detection processing method and system
CN104408724A (en) Depth information method and system for monitoring liquid level and recognizing working condition of foam flotation
CN102252859B (en) Road train straight-line running transverse stability automatic identification system
CN113011331B (en) Method and device for detecting whether motor vehicle gives way to pedestrians, electronic equipment and medium
CN106156752A (en) A kind of model recognizing method based on inverse projection three-view diagram
CN104794906A (en) Vehicle management platform of outdoor parking lot exit
CN110929676A (en) Deep learning-based real-time detection method for illegal turning around
CN103837087A (en) Automatic pantograph detection method based on active shape model
CN113610014A (en) System and method for detecting freight vehicle with shielding number plate exceeding limit
CN108922176A (en) A kind of quick judgment method of mixed traffic conflict state
CN118196105B (en) Road quality detection method based on image recognition
CN116631187A (en) Intelligent acquisition and analysis system for case on-site investigation information
CN117116046A (en) Traffic common event detection method based on single-stage target detection
CN102622582B (en) Road pedestrian event detection method based on video
CN114612731B (en) Intelligent identification method and system for road flatness detection

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant