Real-time vehicle detection and trajectory tracking method in traffic video based on depth learning

Abstract

The invention discloses a real-time vehicle detection and trajectory tracking method in a traffic video, which comprises the following steps: adopting an object detection algorithm based on depth learning to detect the position of a vehicle in the traffic video, and extracting a feature vector and a category of the vehicle; tracking the vehicle and getting the track. The whole scheme of the invention has strong robustness, low missed detection rate, and is easy to be extended to various vehicle categories, so as to meet the requirements of vehicle detection and continuous tracking in the surveillance video.

Description

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

AI Technical Summary

Benefits of technology

This technology helps identify vehicles quickly while also being able to track them accurately over complicated scenarios that may change their movement or have different types of traffic lights attached thereto at varying speeds.

Problems solved by technology

This patented technology describes two main technical problem addressed in this patents relating to vehicle detection systems that use visual targets like license plates or road signs. These conventional techniques require manual effort and slow processing times which limit how well these devices work with complicated scenarios without being able to divide them into smaller groups.

Images