Non-human multi-target real-time track extraction method in traffic video scene

Abstract

The invention discloses a non-human multi-target real-time track extraction method in a traffic video scene. The non-human multi-target real-time track extraction method comprises the steps: carryingout the target detection of an inputted traffic video image through combining with a background difference method, and obtaining all traffic entity targets of each frame in the video image; realizingclassification and identification of motor vehicles, non-motor vehicles and pedestrians by utilizing the length-width ratio and the two-dimensional area characteristics of the traffic entity targets;designing a non-human multi-target real-time trajectory matching tracking algorithm, matching the trajectories of the traffic entity targets detected in each frame with the trajectories of the previously detected traffic entity targets one by one, and classifying the traffic entity targets to which the traffic entity targets belong to realize tracking if the matching succeeds; and extracting the trajectory of the traffic entity target subjected to trajectory matching tracking, and determining the driving direction of the traffic entity through judgment of starting and ending coordinate positions of the target trajectory. According to the invention, the extracted multi-target trajectory can be automatically matched to different traffic entities and different traffic flow directions of non-human and non-human, and the robustness is good.

Description

technical field [0001] The present invention relates to the technical field of intelligent transportation, and more specifically relates to a real-time trajectory extraction method based on traffic video scenes. Background technique [0002] In actual mixed traffic scenarios, the real-time trajectory data of machines and non-humans is the basis for intelligent active safety control technologies such as early warning of traffic dynamic conflicts. Based on the existing traffic video monitoring network, it is one of the representative and practical methods in the traffic field to analyze the acquired traffic video images, extract the trajectory of the machine and non-human in real time, and then carry out conflict warning and other applications. . However, the existing research focuses on the single application of vehicle detection and tracking, and its solutions rarely focus on the practical application in traffic scenarios, and seldom study the overall solution of real-time ...

AI Technical Summary

Problems solved by technology

However, the existing research focuses on the single application of vehicle detection and tracking, and its solutions rarely pay attention to the practical application in traffic scenarios, and rarely study the overall solution of real-time trajectory extraction of machine, non-human and multi-target from the perspective of the entire actual traffic scene application.

[0003]There are some difficulties and challenges in the real-time trajectory extraction of off-board non-human in traffic video scenes, such as low real-time performance and low practical level in trajectory extraction

At present, traffic video trajectory extraction mostly needs to be manually marked based on video image frames and can only extract the trajectory of one vehicle at the same time, such as Geroge, Tracker, and NGSIM software, which cannot be applied to actual traffic scenes where motor vehicles, non-motor vehicles, and pedestrians are mixed.

Images