CN112102609A - Intelligent detection system for highway vehicle violation behaviors - Google Patents
Intelligent detection system for highway vehicle violation behaviors Download PDFInfo
- Publication number
- CN112102609A CN112102609A CN202010160392.3A CN202010160392A CN112102609A CN 112102609 A CN112102609 A CN 112102609A CN 202010160392 A CN202010160392 A CN 202010160392A CN 112102609 A CN112102609 A CN 112102609A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- image
- license plate
- deep learning
- detection
- 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.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/54—Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Analytical Chemistry (AREA)
- Chemical & Material Sciences (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to an intelligent detection system for highway vehicle violation behaviors. And detecting and identifying the vehicles, the steering lamps, the license plates and the license plate numbers in the video images by adopting a target detection algorithm based on deep learning. And respectively judging the illegal behaviors that the vehicle occupies an emergency lane, exceeds the speed and does not turn on a turn light when entering a ramp by utilizing a zoning method, an overspeed detection method and a vehicle turn light flashing state identification algorithm. And storing the video band for the illegal vehicle to run and the license plate information of the vehicle in a server. The invention can accurately detect the illegal behaviors that the vehicles on the expressway occupy emergency lanes, go over speed and enter ramps without turning lamps in real time, and provides important guarantee for standardizing the vehicles to run according to the traffic regulations and ensuring the traveling safety of drivers. The system is simple to operate, can automatically detect the violation behaviors, reduces the labor cost, and is suitable for large-area popularization and use.
Description
Technical Field
The invention relates to the field of traffic supervision, in particular to an intelligent detection system for highway vehicle violation behaviors.
Background
In recent years, with the increasing urban population, the demand of motor vehicles is increasing, thus bringing a series of traffic problems. Among them, a traffic accident caused by the driver not driving the vehicle according to the norm is particularly serious. Therefore, the regulation violation correcting work is done, the legal rights and interests of citizens are protected, traffic accidents are reduced, the road traffic order is maintained, and the traffic safety and the traffic smoothness are protected.
It is more hazardous to drive a vehicle against a violation on a highway. When the vehicle has the behaviors of speeding, occupying an emergency lane, entering a ramp and not driving a steering lamp and the like, rear-end collision of the vehicle is likely to be caused, and traffic accidents are caused. Therefore, the traffic accident can be effectively prevented by detecting the violation behaviors of the vehicles running on the highway.
At present, the detection of the violation behaviors of the vehicles on the highway is generally carried out by manually screening the violation behaviors according to the video acquisition result and then storing the video of the violation vehicles and the vehicle information. This method has the following problems in practical use:
1. vehicles in the video need to be monitored continuously for 24 hours manually, and the data volume is large, and timeliness and accuracy are difficult to guarantee;
2. the screening process of the video and the storage of the vehicle information are all completed manually, the operation quality is difficult to ensure, and subjective errors are possibly introduced;
3. the violation vehicles are monitored manually, so that the time cost and the labor cost are high, and the current complex traffic condition cannot be adapted;
therefore, the development and the development of the intelligent transportation system based on the effective combination of the technologies such as electronic control, computer processing, artificial intelligence and the like have important research value and significance.
Disclosure of Invention
The invention aims to provide an intelligent detection system for the violation behaviors of the vehicles on the highway, aiming at the defects of high cost, poor timeliness and accuracy and the like in the prior art, and the intelligent detection system has the advantages of manpower cost saving, high stability, simplicity in operation and good instantaneity. In order to achieve the purpose, the invention designs the intelligent detection system for the violation behaviors of the vehicles on the expressway, which can accurately detect the violation behaviors that the vehicles on the expressway occupy an emergency lane, run at an overspeed and enter a ramp without turning on a turn light in real time;
the technical scheme adopted by the invention for realizing the purpose is as follows:
an intelligent detection system for highway vehicle violation, comprising:
the video decoding program is used for reading the highway vehicle video stream through the camera, converting the highway vehicle video stream into an image and cutting the image by utilizing coordinate transformation;
the expressway scene regional algorithm program is used for identifying the image cut by the video decoding program as an emergency lane region, a lower lane entrance region or a speed measuring region according to the coordinates of the vehicle;
the target detection program is used for detecting the vehicle, the vehicle lamp, the license plate and the license plate number in the image cut by the video decoding program;
the overspeed detection algorithm program is used for storing an overspeed detection algorithm, calculating the speed of the vehicle in the speed measurement area in real time by using the overspeed detection algorithm, and judging whether the vehicle has overspeed behavior according to a set speed limit value;
and the vehicle steering lamp flickering state recognition algorithm program is used for storing a steering lamp flickering recognition algorithm and judging whether the steering lamp is turned on or not when the vehicle enters the lower crossing according to the vehicle lamp color change of the vehicle in the lower crossing area by using the steering lamp flickering recognition algorithm.
The object detection program comprises: the vehicle license plate detection system comprises a vehicle detection model based on deep learning, a vehicle lamp detection model based on deep learning, a license plate detection model based on deep learning and a license plate number detection model based on deep learning.
The input of the vehicle detection model based on the deep learning is the image which is cut by the video decoding program and is output as the detected vehicle image.
The input of the car light detection model based on the deep learning is a detected car image output by the car detection model based on the deep learning, and the output is a car light image of the car.
The input of the license plate detection model based on the deep learning is a detected vehicle image output by the vehicle detection model based on the deep learning, and the output is a license plate image of the vehicle.
The input of the license plate number detection model based on deep learning is the license plate image of the vehicle output by the license plate detection model based on deep learning, and the output is the license plate number of the vehicle.
An intelligent detection method for highway vehicle violation behaviors comprises the following steps:
1) the video decoding program reads the highway vehicle video stream through the camera, converts the highway vehicle video stream into an image and cuts the image by utilizing coordinate transformation;
2) dividing the cut image into an emergency lane area, a lower lane exit area and a speed measuring area according to the coordinates of the vehicle by the expressway scene zoning algorithm program;
3) inputting the decoded and cut images into a vehicle detection model based on deep learning by an object detection program for vehicle detection, outputting the detected vehicle images, and executing steps 4.1) to 4.3);
4.1) judging whether a vehicle exists in the emergency lane area according to the detected coordinates of the vehicle image, if so, determining that the vehicle occupies the emergency lane, and executing the step 7); otherwise, returning to the step 3);
4.2) judging whether a vehicle exists in the lower intersection area according to the coordinates of the detected vehicle image, if so, inputting the detected vehicle image into a vehicle lamp detection model based on deep learning to perform vehicle lamp detection, and executing the step 5); otherwise, returning to the step 3);
4.3) judging whether a vehicle exists in the speed measuring area according to the detected coordinates of the vehicle image, and if so, executing the step 6); otherwise, returning to the step 3);
5) the vehicle steering lamp flickering state recognition algorithm program detects the steering lamp turning behavior of the vehicle in the lower crossing area by using a vehicle steering lamp flickering recognition algorithm, judges whether the vehicle turns on the steering lamp, and if so, executes the step 7); otherwise, returning to the step 4.2);
6) judging whether the vehicle has overspeed behavior by using an overspeed detection algorithm program and executing step 7) if the vehicle has overspeed behavior; otherwise, returning to the step 4.3);
7) inputting the detected vehicle image into a license plate detection model based on deep learning to perform license plate detection, and outputting a license plate image of the vehicle;
8) and inputting the license plate image of the vehicle into a license plate number detection model based on deep learning to detect the license plate number, and outputting the license plate number of the vehicle.
The vehicle steering lamp flicker identification algorithm comprises the following steps:
converting the detected car light image from an RGB color space into an HSV color space, wherein the hue H is measured by using an angle, the value range is 0-360 degrees, different angles represent different colors, the color of the vehicle steering lamp is judged by using hue characteristics, the steering lamp is amber in a bright state, and whether the vehicle steering lamp is in a bright state or not is judged according to the hue range of the amber; when the vehicle steering lamp is in a turned-off state, the color tone of the steering lamp changes and deviates from the amber range, whether the vehicle turns on the steering lamp is judged according to the continuous N frames of vehicle lamp images in the lower intersection area, and if the steering lamp changes from turning on to turning off or from turning off to turning on, the vehicle turns on the steering lamp; otherwise, the vehicle has violation operation.
The overspeed detection algorithm is as follows:
the moving distance of the central point of the same vehicle in two adjacent frames is used as the reference pixel distance for the vehicle to run, and the central position coordinates of the vehicle target frames in the front frame and the rear frame are obtained according to the vehicle target frames detected in the front frame and the rear frame; setting the coordinates of the upper left corner and the lower right corner of the vehicle target frame in the previous frame as follows: (x1, y1), (x2, y2), the coordinates of the upper left corner and the lower right corner of the vehicle target frame of the next frame are: (x3, y3) and (x4, y4), the front and rear frame vehicle target frame center coordinates are (x5, y5) and (x6, y6), respectively; the central coordinates of the front and rear frames of the vehicle target frame are respectively expressed as:
and
if the frame frequency of the video stream is F and the actual Width of the vehicle is Width, the vehicle speed V can be expressed as:
and judging whether the vehicle has overspeed behavior according to the speed threshold value of the vehicle running on the highway.
The invention has the following beneficial effects and advantages:
1. the invention adopts a target detection algorithm based on deep learning and trains a detection model on a special highway vehicle data set, thereby effectively improving the accuracy of detecting vehicles, lamps, license plates and license plate numbers and simultaneously ensuring the real-time property of a detection system.
2. The invention can automatically complete the detection of the illegal behaviors that the vehicle occupies an emergency lane, exceeds the speed and enters a ramp without turning on the steering lamp by utilizing a scene zoning algorithm of the expressway, an overspeed detection algorithm and a vehicle steering lamp flickering state recognition algorithm. And automatically storing license plate information of the violation vehicle and the video band of vehicle running into the server.
3. The method has the advantages of low cost, simple operation and good stability, can realize intelligent detection of the violation behaviors of the vehicles on the highway, and is suitable for large-area popularization.
Drawings
FIG. 1 is a flow chart of the algorithm of the present invention;
FIG. 2 is a flow chart of a vehicle turn signal flashing state identification algorithm of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the accompanying drawings and examples.
The intelligent detection system is internally provided with: and detecting models and trained model weight parameters based on deeply learned vehicles, vehicle lamps, license plates and license plate numbers.
The intelligent detection system is internally provided with: a video decoding procedure and an object detection procedure.
The intelligent detection system is internally provided with: and (4) an expressway scene zoning algorithm program.
The intelligent detection system is internally provided with: and (4) overspeed detection program.
The intelligent detection system is internally provided with: and identifying the flickering state of the vehicle steering lamp.
The detection model installed in the intelligent detection system is a target detection model based on deep learning, and comprises the following steps: the vehicle license plate number recognition system comprises a vehicle detection model based on deep learning, a vehicle lamp detection model based on deep learning, a license plate detection model based on deep learning and a license plate number recognition model based on deep learning.
The vehicle detection model based on deep learning, the vehicle lamp detection model based on deep learning, the license plate detection model based on deep learning and the license plate number identification model based on deep learning all adopt neural network models.
The model weight parameters installed in the intelligent detection system are obtained by training the detection model by utilizing a special highway vehicle data set, a special vehicle lamp data set and a special license plate data set.
The video decoding program installed in the intelligent detection system comprises reading of a video stream, converting the video stream into an image and cutting the image into a size suitable for the input of the detection model by using coordinate transformation.
The target detection algorithm program installed in the intelligent detection system connects the deep learning models of the vehicle, the vehicle lamp, the license plate and the license plate number in series. The input of the vehicle detection model based on the deep learning is the image after the video decoding program decodes and cuts, and the image is output as the detected vehicle image. The input of the deep learning-based vehicle lamp detection model is the detected vehicle image output by the deep learning-based vehicle detection model, and the output is the vehicle lamp image of the vehicle. The input of the license plate detection model based on the deep learning is a detected vehicle image output by the vehicle detection model based on the deep learning, and the output is a license plate image of the vehicle. The input of the license plate number detection model based on deep learning is the license plate image of the vehicle output by the license plate detection model based on deep learning, and the output is the license plate number of the vehicle.
The expressway scene zoning algorithm program installed in the intelligent detection system divides the expressway scene into an emergency lane zone, a lower lane zone and a speed measuring zone.
The overspeed detection program installed in the intelligent detection system can calculate the speed of the running vehicle in real time and judge whether the vehicle has overspeed behavior according to the set speed limit value.
The vehicle turn light flickering state recognition program installed in the intelligent detection system can judge whether the vehicle turns on the turn light or not when entering the ramp port according to the change of the color of the vehicle light.
The automatic detection flow of the violation behaviors of the present invention is described in detail below with reference to fig. 1.
The invention adopts a target detection model based on deep learning, which respectively comprises the following steps: the vehicle license plate number recognition system comprises a vehicle detection model based on deep learning, a vehicle lamp detection model based on deep learning, a license plate detection model based on deep learning and a license plate number recognition model based on deep learning. And respectively training the detection models by using a special expressway vehicle data set, a special expressway vehicle light data set and a special expressway license plate data set, and storing the trained model weight parameters in the main controller for online detection of the vehicle, the vehicle lights and the license plate.
Firstly, an intelligent detection system carries out video decoding on a video of a vehicle running on a highway, and a decoded image is divided into an emergency lane area, a lower lane entrance area and a speed measuring area. And then loading the decoded image to a vehicle detection model based on deep learning for vehicle detection, judging whether a vehicle exists in the emergency crossing area according to the detected coordinates of the vehicle, and if so, indicating that the vehicle occupies an emergency lane. Judging whether vehicles exist in the lower crossing area or not, if so, cutting an original image according to the obtained vehicle coordinates, loading the cut vehicle image to a vehicle lamp detection model based on deep learning, and detecting the behavior of turning on the steering lamp of the vehicle at the lower crossing by using a vehicle steering lamp flicker recognition algorithm. And judging whether the vehicle has overspeed behavior or not by using an overspeed detection algorithm for the vehicle in the speed measurement area. If the vehicle has illegal behaviors of occupying an emergency lane, not driving a steering lamp on a lower ramp and speeding, the system loads a vehicle image to a license plate detection model based on deep learning, cuts the vehicle image according to the obtained license plate coordinates, and loads the cut license plate image to a license plate number recognition model based on the deep learning to finish the recognition of the vehicle license plate number. And finally, storing the video band for the illegal vehicle to run and the license plate number of the vehicle in a server.
The turn signal flicker recognition algorithm of fig. 1 is described in detail below with reference to fig. 2.
The detected car light image is a color image composed of three channels of RGB. In the RGB color space, the color is determined by the three channels RGB at the same time. In order to simplify the color model, the detected image is firstly converted from the RGB color space into the HSV color space, and the color of the vehicle steering lamp is judged by utilizing the hue characteristic in the HSV color space. The turn signal lamp is amber in the bright state, so that whether the turn signal lamp of the vehicle is in the bright state can be judged according to the hue range of the amber. When the turn signal of the vehicle is in an off state, the color tone of the turn signal is deviated. In order to ensure that the flicker state of the steering lamp can be accurately judged, the method judges the vehicle lamp detection images of N continuous frames in the lower intersection area, and when the steering lamp changes from on to off or from off to on, the vehicle turns on the steering lamp. Otherwise, if the vehicle has violation operation, the system converts the image frame of the vehicle in the range of the lower ramp into a video segment and stores the video segment and the license plate information of the vehicle in the server, wherein n is the number of frames.
The overspeed detection algorithm of FIG. 2 is described in detail below
The invention adopts the distance moved by the central point of the same vehicle in the front and rear frames as the reference pixel distance for the vehicle to run. Parameters of the speed calculation inputs include: position information of two frames of vehicle target frames, frame frequency of videos and actual width information of vehicles.
The distance of the movement of the central point of the same vehicle in the adjacent front and rear frames is used as the distance of the reference pixel of the vehicle running, and the central position coordinates of the vehicle frames in the front and rear frames can be obtained according to the vehicle target frames detected in the front and rear frames. Setting the coordinates of the upper left corner and the lower right corner of the front frame and the rear frame of the vehicle target frame as follows: (x1, y1), (x2, y2), (x3, y3), and (x4, y4), and the vehicle center coordinates are (x5, y5), and (x6, y6), respectively. The coordinates of the center point of the two front and rear frames of the vehicle can be expressed as:
and
the frame frequency of the video stream is F, the actual Width of the vehicle is Width, and the vehicle speed V can be expressed as
And judging whether the vehicle has overspeed behavior according to the speed threshold value of the vehicle running on the highway. If the violation behaviors exist, the system converts the image frame of the vehicle in the range of the lower ramp into a video band and stores the video band and the license plate information of the vehicle in the server.
Claims (9)
1. An intelligent detection system for highway vehicle violation behaviors, comprising:
the video decoding program is used for reading the highway vehicle video stream through the camera, converting the highway vehicle video stream into an image and cutting the image by utilizing coordinate transformation;
the expressway scene regional algorithm program is used for identifying the image cut by the video decoding program as an emergency lane region, a lower lane entrance region or a speed measuring region according to the coordinates of the vehicle;
the target detection program is used for detecting the vehicle, the vehicle lamp, the license plate and the license plate number in the image cut by the video decoding program;
the overspeed detection algorithm program is used for storing an overspeed detection algorithm, calculating the speed of the vehicle in the speed measurement area in real time by using the overspeed detection algorithm, and judging whether the vehicle has overspeed behavior according to a set speed limit value;
and the vehicle steering lamp flickering state recognition algorithm program is used for storing a steering lamp flickering recognition algorithm and judging whether the steering lamp is turned on or not when the vehicle enters the lower crossing according to the vehicle lamp color change of the vehicle in the lower crossing area by using the steering lamp flickering recognition algorithm.
2. The intelligent detection system for highway vehicle violation according to claim 1, wherein said object detection program comprises: the vehicle license plate detection system comprises a vehicle detection model based on deep learning, a vehicle lamp detection model based on deep learning, a license plate detection model based on deep learning and a license plate number detection model based on deep learning.
3. The intelligent detection system for the highway vehicle violation behavior of claim 2 wherein the deep learning based vehicle detection model has an input of a video decoding program for decoding the cropped image and an output of a detected vehicle image.
4. The intelligent detection system for highway vehicle violation according to claim 2 wherein the deep learning based vehicle light detection model has the input of the detected vehicle image output by the deep learning based vehicle detection model and the output of the detected vehicle image as the vehicle light image.
5. The intelligent detection system for highway vehicle violation behaviors of claim 2, wherein the input of said deep learning based license plate detection model is the detected vehicle image output by the deep learning based vehicle detection model and the output is the license plate image of the vehicle.
6. The intelligent detection system for the violation behaviors of the highway vehicle as recited in claim 2, wherein the input of the deep learning-based license plate number detection model is a license plate image of the vehicle output by the deep learning-based license plate detection model, and the output is a license plate number of the vehicle.
7. An intelligent detection method for highway vehicle violation behaviors is characterized by comprising the following steps:
1) the video decoding program reads the highway vehicle video stream through the camera, converts the highway vehicle video stream into an image and cuts the image by utilizing coordinate transformation;
2) dividing the cut image into an emergency lane area, a lower lane exit area and a speed measuring area according to the coordinates of the vehicle by the expressway scene zoning algorithm program;
3) inputting the decoded and cut images into a vehicle detection model based on deep learning by an object detection program for vehicle detection, outputting the detected vehicle images, and executing steps 4.1) to 4.3);
4.1) judging whether a vehicle exists in the emergency lane area according to the detected coordinates of the vehicle image, if so, determining that the vehicle occupies the emergency lane, and executing the step 7); otherwise, returning to the step 3);
4.2) judging whether a vehicle exists in the lower intersection area according to the coordinates of the detected vehicle image, if so, inputting the detected vehicle image into a vehicle lamp detection model based on deep learning to perform vehicle lamp detection, and executing the step 5); otherwise, returning to the step 3);
4.3) judging whether a vehicle exists in the speed measuring area according to the detected coordinates of the vehicle image, and if so, executing the step 6); otherwise, returning to the step 3);
5) the vehicle steering lamp flickering state recognition algorithm program detects the steering lamp turning behavior of the vehicle in the lower crossing area by using a vehicle steering lamp flickering recognition algorithm, judges whether the vehicle turns on the steering lamp, and if so, executes the step 7); otherwise, returning to the step 4.2);
6) judging whether the vehicle has overspeed behavior by using an overspeed detection algorithm program and executing step 7) if the vehicle has overspeed behavior; otherwise, returning to the step 4.3);
7) inputting the detected vehicle image into a license plate detection model based on deep learning to perform license plate detection, and outputting a license plate image of the vehicle;
8) and inputting the license plate image of the vehicle into a license plate number detection model based on deep learning to detect the license plate number, and outputting the license plate number of the vehicle.
8. The intelligent detection method for the violation of highway vehicles according to claim 7, wherein said vehicle turn signal flash recognition algorithm is:
converting the detected car light image from an RGB color space into an HSV color space, wherein the hue H is measured by using an angle, the value range is 0-360 degrees, different angles represent different colors, the color of the vehicle steering lamp is judged by using hue characteristics, the steering lamp is amber in a bright state, and whether the vehicle steering lamp is in a bright state or not is judged according to the hue range of the amber; when the vehicle steering lamp is in a turned-off state, the color tone of the steering lamp changes and deviates from the amber range, whether the vehicle turns on the steering lamp is judged according to the continuous N frames of vehicle lamp images in the lower intersection area, and if the steering lamp changes from turning on to turning off or from turning off to turning on, the vehicle turns on the steering lamp; otherwise, the vehicle has violation operation.
9. The intelligent detection method for highway vehicle violation according to claim 7, wherein said overspeed detection algorithm is:
the moving distance of the central point of the same vehicle in two adjacent frames is used as the reference pixel distance for the vehicle to run, and the central position coordinates of the vehicle target frames in the front frame and the rear frame are obtained according to the vehicle target frames detected in the front frame and the rear frame; setting the coordinates of the upper left corner and the lower right corner of the vehicle target frame in the previous frame as follows: (x1, y1), (x2, y2), the coordinates of the upper left corner and the lower right corner of the vehicle target frame of the next frame are: (x3, y3) and (x4, y4), the front and rear frame vehicle target frame center coordinates are (x5, y5) and (x6, y6), respectively; the central coordinates of the front and rear frames of the vehicle target frame are respectively expressed as:
and
if the frame frequency of the video stream is F and the actual Width of the vehicle is Width, the vehicle speed V can be expressed as:
and judging whether the vehicle has overspeed behavior according to the speed threshold value of the vehicle running on the highway.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010160392.3A CN112102609A (en) | 2020-03-10 | 2020-03-10 | Intelligent detection system for highway vehicle violation behaviors |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010160392.3A CN112102609A (en) | 2020-03-10 | 2020-03-10 | Intelligent detection system for highway vehicle violation behaviors |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112102609A true CN112102609A (en) | 2020-12-18 |
Family
ID=73750461
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010160392.3A Pending CN112102609A (en) | 2020-03-10 | 2020-03-10 | Intelligent detection system for highway vehicle violation behaviors |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112102609A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113111876A (en) * | 2021-04-14 | 2021-07-13 | 深圳市旗扬特种装备技术工程有限公司 | Method and system for obtaining evidence of traffic violation |
| CN113380047A (en) * | 2021-06-22 | 2021-09-10 | 重庆盛海科技发展有限公司 | Illegal parking detection method based on traditional camera |
| CN114758511A (en) * | 2022-06-14 | 2022-07-15 | 深圳市城市交通规划设计研究中心股份有限公司 | Sports car overspeed detection system, method, electronic equipment and storage medium |
| TWI808603B (en) * | 2022-01-10 | 2023-07-11 | 中華電信股份有限公司 | A ramp exit road condition detection system, method, and computer-readable medium thereof |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105608906A (en) * | 2016-03-29 | 2016-05-25 | 成都理工大学 | System for monitoring illegal emergency lane occupancy of expressway motor vehicles and implementation method |
| KR101717613B1 (en) * | 2016-12-27 | 2017-03-17 | 주식회사한맥아이피에스 | The moving vehicle detection system using an object tracking algorithm based on edge information, and method thereof |
| CN108091142A (en) * | 2017-12-12 | 2018-05-29 | 公安部交通管理科学研究所 | For vehicle illegal activities Tracking Recognition under highway large scene and the method captured automatically |
| CN209103529U (en) * | 2018-11-16 | 2019-07-12 | 重庆盛海科技发展有限公司 | Intelligent transportation violation snap-shooting system |
| CN110136447A (en) * | 2019-05-23 | 2019-08-16 | 杭州诚道科技股份有限公司 | Lane change of driving a vehicle detects and method for distinguishing is known in illegal lane change |
| CN110136449A (en) * | 2019-06-17 | 2019-08-16 | 珠海华园信息技术有限公司 | Traffic video frequency vehicle based on deep learning disobeys the method for stopping automatic identification candid photograph |
| KR20190120968A (en) * | 2018-04-17 | 2019-10-25 | 동명대학교산학협력단 | Traffic violation enforcement system using drones |
-
2020
- 2020-03-10 CN CN202010160392.3A patent/CN112102609A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105608906A (en) * | 2016-03-29 | 2016-05-25 | 成都理工大学 | System for monitoring illegal emergency lane occupancy of expressway motor vehicles and implementation method |
| KR101717613B1 (en) * | 2016-12-27 | 2017-03-17 | 주식회사한맥아이피에스 | The moving vehicle detection system using an object tracking algorithm based on edge information, and method thereof |
| CN108091142A (en) * | 2017-12-12 | 2018-05-29 | 公安部交通管理科学研究所 | For vehicle illegal activities Tracking Recognition under highway large scene and the method captured automatically |
| KR20190120968A (en) * | 2018-04-17 | 2019-10-25 | 동명대학교산학협력단 | Traffic violation enforcement system using drones |
| CN209103529U (en) * | 2018-11-16 | 2019-07-12 | 重庆盛海科技发展有限公司 | Intelligent transportation violation snap-shooting system |
| CN110136447A (en) * | 2019-05-23 | 2019-08-16 | 杭州诚道科技股份有限公司 | Lane change of driving a vehicle detects and method for distinguishing is known in illegal lane change |
| CN110136449A (en) * | 2019-06-17 | 2019-08-16 | 珠海华园信息技术有限公司 | Traffic video frequency vehicle based on deep learning disobeys the method for stopping automatic identification candid photograph |
Non-Patent Citations (3)
| Title |
|---|
| 孙学聪等: "一种简单的基于图像处理交通信号灯识别算法", 《汽车实用技术》 * |
| 山东省交通安全教育学校指导委员会: "《机动车驾驶员交通安全教育教程》", 31 May 2005 * |
| 游四海等编: "《汽车服务工程》", 30 June 2005 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113111876A (en) * | 2021-04-14 | 2021-07-13 | 深圳市旗扬特种装备技术工程有限公司 | Method and system for obtaining evidence of traffic violation |
| CN113380047A (en) * | 2021-06-22 | 2021-09-10 | 重庆盛海科技发展有限公司 | Illegal parking detection method based on traditional camera |
| TWI808603B (en) * | 2022-01-10 | 2023-07-11 | 中華電信股份有限公司 | A ramp exit road condition detection system, method, and computer-readable medium thereof |
| CN114758511A (en) * | 2022-06-14 | 2022-07-15 | 深圳市城市交通规划设计研究中心股份有限公司 | Sports car overspeed detection system, method, electronic equipment and storage medium |
| CN114758511B (en) * | 2022-06-14 | 2022-11-25 | 深圳市城市交通规划设计研究中心股份有限公司 | Sports car overspeed detection system, method, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112102609A (en) | Intelligent detection system for highway vehicle violation behaviors | |
| CN110197589B (en) | Deep learning-based red light violation detection method | |
| CN111815959B (en) | Vehicle violation detection method and device and computer readable storage medium | |
| CN111212772B (en) | Method and device for determining a driving strategy of a vehicle | |
| EP1930863B1 (en) | Detecting and recognizing traffic signs | |
| CN103778786B (en) | A kind of break in traffic rules and regulations detection method based on remarkable vehicle part model | |
| CN102768801B (en) | Method for detecting motor vehicle green light follow-up traffic violation based on video | |
| CN109299674B (en) | Tunnel illegal lane change detection method based on car lamp | |
| CN113112830B (en) | Signal control intersection emptying method and system based on laser radar and track prediction | |
| CN108407806A (en) | Auxiliary driving method and device | |
| CN103832357B (en) | A kind of lane-departure warning system and method based on machine vision | |
| CN104228670B (en) | The implementation method of a kind of drive recorder for oversize vehicle | |
| CN111462491A (en) | Expressway confluence area traffic conflict early warning method based on ramp control | |
| CN113033275B (en) | Vehicle lane-changing non-turn signal lamp analysis system based on deep learning | |
| CN111649918A (en) | Method for monitoring turning on of high beam and continuous tracking of vehicle in real time | |
| CN112606831A (en) | Anti-collision warning information external interaction method and system for passenger car | |
| CN115782905B (en) | Automatic driving safety degree quantification system for driving vehicle | |
| CN111967384A (en) | Vehicle information processing method, device, equipment and computer readable storage medium | |
| CN205788764U (en) | The automatic evidence-collecting device of steering indicating light is not opened in a kind of vehicle modified line | |
| US20230401875A1 (en) | Method for recognizing illumination state of traffic lights, electronic device, and storage medium | |
| CN114693722B (en) | Vehicle driving behavior detection method, detection device and detection equipment | |
| CN116543570A (en) | Traffic light intelligent control system based on PLC | |
| CN112101230B (en) | Method and system for detecting opening of head lamp of road tunnel passing vehicle | |
| CN113177443A (en) | Method for intelligently identifying road traffic violation based on image vision | |
| CN202694585U (en) | Vehicle-mounted visual traffic signal lamp intelligent detection apparatus |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20201218 |
|
| RJ01 | Rejection of invention patent application after publication |