A Method of Object Tracking

Abstract

The invention provides a method for tracking targets. The method includes detecting an initial tracking area; initializing the system state, establishing a system recursive state model with Gaussian noise and establishing a system state model; reading in a new frame from a video, and graying and vectorizing the new frame; performing sampling to form a t-moment particle state xt, using a principal component analysis method, extracting characteristics from the t-moment particle state, and calculating the distance of the characteristics and weight of each particle in a particle group; calculating an error of mean square of the group of particles according to the weight of each particle, determining whether to perform re-sampling according to the error of mean square, selecting optimal particles to output if re-sampling is not performed, and displaying an optimal tracking result in the video. According to the method for tracking targets, particle degradation and depletion caused by excessive re-sampling are avoided, and tracking accuracy and robustness are improved. The method is applicable to the target tracking, the target tracking of multi-obstacles and face tracking under the complicated background in video or sequential images.

Description

technical field [0001] The present invention relates to the field of images, more specifically, to the field of target tracking. Background technique [0002] Target tracking technology under complex background is one of the core research topics in the field of computer vision, which includes computer graphics, target recognition, artificial intelligence and automatic control and other technologies. Various current tracking algorithms, such as optical flow method and Kalman filter method, cannot meet the requirements of tracking accuracy, while particle filter method is better than other tracking methods in solving nonlinear and non-Gaussian problems. However, the basic particle filter algorithm requires continuous resampling, which eventually leads to particle degradation and sample depletion, and it is difficult to meet the target tracking requirements; moreover, the basic particle filter algorithm cannot achieve target tracking in complex backgrounds. In order to apply t...

AI Technical Summary

Problems solved by technology

[0006] Based on the research status at home and abroad, there are still many deficiencies in target tracking based on particle filter, mainly reflected in the fact that these achievements cannot solve target tracking in complex backgrounds well, and the tracking accuracy and robustness cannot meet the requirements.

Practical target tracking is mostly in complex environments, and the current particle filter method does not have a good way to deal with complex environments. The current main research is still on optimizing the system state and optimizing the resampling method

Images