Multi-target tracking method, system and device based on optical flow and Kalman filtering

Abstract

The invention belongs to the field of intelligent driving, particularly relates to a multi-target tracking method, system and device based on optical flow and Kalman filtering, and aims to solve the problem that multi-target tracking excessively depends on detection and is high in matching difficulty. The method of the system comprises the following steps: acquiring an image frame at t moment in an input video; performing multi-target detection on the image frame, and constructing a target region set of detection targets; according to the target tracked at time t-1, predicting the corresponding target area of the target in the image frame at time t by the preset tracking model, and constructing the target area set of the predicted target; for the target area sets of the detection target and the prediction target, obtaining a target area matching relationship in the two target area sets through area similarity and a Hungary algorithm; and taking the successfully matched target area in the target area set of the detection target as the target area tracked at the t moment. According to the invention, accurate prediction of multiple targets is realized, and the accuracy of position matching and tracking of multiple frames of targets is improved.

Description

technical field [0001] The invention belongs to the field of intelligent driving, and in particular relates to a multi-target tracking method, system and device based on optical flow and Kalman filtering. Background technique [0002] Multi-target tracking is an important part of intelligent driving perception. By identifying and tracking objects around the vehicle, including vehicles, pedestrians and traffic signs, etc., it provides information for the intelligent driving back-end control unit. [0003] Most of the existing multi-target tracking methods are tracking methods based on target detection, which correlate and match the detection results with the previously tracked targets to obtain the final target tracking track. In the face of various weather conditions and complex road environments, the image-based multi-target detection and tracking method is an effective and low-cost method, especially the multi-target tracking method based on Kalman filtering, which is fast...

AI Technical Summary

Problems solved by technology

[0004] In order to solve the above-mentioned problems in the prior art, that is, in order to solve the problem that multi-target tracking is too dependent on detection results and difficult to match, the first aspect of the present invention proposes a multi-target tracking method based on optical flow and Kalman filtering , the method includes:

Images