Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Deep learning-based red light violation detection method

A detection method and deep learning technology, which is applied in the field of red light violation detection at traffic light intersections, can solve the problems of high cost, inflexibility, and low accuracy, and achieve the effect of high accuracy and low cost

Active Publication Date: 2019-09-03
HANGZHOU TRUSTWAY TECH
View PDF8 Cites 17 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to overcome the disadvantages of high cost, low efficiency, inflexibility, and low accuracy of the above two traditional illegal vehicle monitoring methods, and provide a detection method based on deep learning for running a red light to meet the needs of urban road traffic management. The need for accuracy and flexibility in illegal detection

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Deep learning-based red light violation detection method
  • Deep learning-based red light violation detection method
  • Deep learning-based red light violation detection method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0083] In this embodiment, its basic detection process is as described in S1-S7 above, and the effect of its actual application will be specifically shown below.

[0084] Step S1. Locate the area of ​​the signal light and the area inside and outside the line.

[0085] The signal light area is the position of the signal light in the figure. The signal light area is a regular figure, represented by four counterclockwise endpoints. The area inside the line is composed of inner lanes. The number of lanes is determined by the intersection. The inner lanes are divided into straight lanes and left turn lanes. Lanes and right-turn lanes, lanes consist of lane lines defined by two endpoints in the diagram.

[0086] Since the position of the camera is fixed, the position of the signal light area and the lane line in the transmitted picture frame set K remains unchanged in each picture, so only one positioning of the signal light area, the inside area and the outside area is required. T...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a deep learning-based red light violation detection method, which relates to the field of traffic violation detection. The positions of lanes and traffic lights at a traffic light intersection are determined in advance, and the traffic lights are matched with the corresponding lanes. The condition at the traffic light intersection is photographed and recorded through a camera, a photo set transmitted by the camera later is processed centrally, and whether a vehicle has a red light violation behavior is judged based on the positions of the same vehicle under the same traffic light on different photo frames. Besides, as for an off-line vehicle that is unrecognizable due to a long distance, whether to be a known vehicle is recognized in a vehicle similarity comparisonmethod. Various algorithms and models are combined, and the novel traffic violation detection method meets the requirements of violation detection accuracy and flexibility by urban road traffic management.

Description

technical field [0001] The invention relates to the field of traffic violation detection, in particular to a method for detecting violations of traffic violations at traffic light intersections based on deep learning convolutional neural network image recognition and image similarity comparison. Background technique [0002] With the development of society and economy, cars are playing an increasingly important role in our lives. While facilitating people's daily life, it brings a series of problems such as traffic violations and congestion. Most of them are caused by motor vehicle drivers breaking the rules or running red lights. [0003] At present, in the traffic management system of cities in my country, there are mainly two ways to capture illegal vehicles: one is to bury induction coils underground, and to set up digital cameras on crossbars to capture images of running red lights. This method has several significant disadvantages of high cost, inconvenient installat...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G08G1/017G08G1/04G06K9/00G06K9/46
CPCG08G1/0175G08G1/04G06V20/54G06V20/40G06V20/52G06V10/50G06V10/56G06V2201/08
Inventor 胡程陈教李万清钱逯胡明中唐建斌洪青阳袁友伟彭瀚夏嘉璐
Owner HANGZHOU TRUSTWAY TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products