Highway traffic incident detection method based on deep learning

A road traffic and event detection technology, applied in the field of road traffic event detection based on deep learning, can solve problems such as poor real-time performance and detection accuracy, inability to meet road traffic and congestion event detection, and achieve the effect of improving accuracy

Active Publication Date: 2022-07-08
SOUTH CHINA UNIV OF TECH
View PDF6 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a road traffic event detection method based on deep learning, to solve the problem that the existing target detection method has poor real-time performance and detection accuracy, and cannot satisfy the detection of traffic and congestion events on the road

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
  • Highway traffic incident detection method based on deep learning
  • Highway traffic incident detection method based on deep learning
  • Highway traffic incident detection method based on deep learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0063] like figure 1 Described, a kind of highway traffic incident detection method based on deep learning, comprises the following steps:

[0064] S1. Obtain a sample data set of road traffic vehicles. The Visdrone2019-DET and UA-DETRAC vehicle datasets are collected as object detection datasets, and the VeRi776 vehicle re-identification dataset is used as the Deepsort characterization extraction dataset.

[0065] S2. Process the sample data set, convert the label type of the target detection data set into a format suitable for YOLOv5, and divide it to obtain a training set and a verification set.

[0066] Since the collected data sets are too large, the vehicle targets in the Visdrone2019-DET data set and the UA-DETRAC data set are screened, and the target detection data sets are sorted and divided, including 16,400 images in the training set and 3,426 images in the validation set. And through the python script batch format conversion of the image, get the YOLOv5 model sui...

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 road traffic incident detection method based on deep learning. The method comprises the following steps: obtaining a road traffic vehicle sample data set; processing the sample data set, and dividing the sample data set to obtain a training set and a verification set; an attention module CBAM is added to the original YOLOv5 model, alpha-CIOULoss is used for replacing the original CIOULoss, and an improved YOLOv5-Improved model detection algorithm is obtained; the processed sample data set is input into a YOLOv5-Improved model, and a. Pt weight file after training is obtained; detecting the video test set by using the weight file to obtain target vehicle result information, and inputting the target vehicle result information into a Deepsort target tracking detection algorithm for tracking to obtain specific coordinate information of the vehicle and a vehicle ID; and inputting the event into a pre-written logic algorithm to judge whether the event is a parking or congestion event or not. By adopting the road traffic event detection method based on deep learning, the problem that an existing target detection method is poor in real-time performance and detection precision can be solved.

Description

technical field [0001] The present invention relates to the technical field of traffic event detection, in particular to a deep learning-based highway traffic event detection method. Background technique [0002] With the rapid development of highways, its daily operation also encounters various problems, such as parking on the highway and congestion problems are the top priority. In the early days, it was mainly achieved by strictly controlling the entrances and exits of the expressway, and the traffic police patrolling the expressway. But this consumes manpower and material resources, and the efficiency is also very low. At present, traffic video surveillance systems are common in the market, but these systems are only responsible for collecting video information and transmitting it to the background control room, and do not have the ability to actively identify abnormal behaviors (such as parking, congestion) on the road, and can only be used as an accident. The later v...

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): G06V20/40G06N3/04G06N3/08
CPCG06N3/08G06N3/045Y02T10/40
Inventor 刘永桂黄家琛
Owner SOUTH CHINA UNIV OF 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
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap