Method and system for real-time tracking of dynamic targets based on improved geometric particle filter

Abstract

The invention discloses a dynamic target real-time tracking method based on improved geometric particle filter, which belongs to the technical field of computer vision. The present invention first obtains the target image, extracts the feature points and edge features of the target image, and performs morphological expansion on the edge features to obtain feature pixels; then acquires the initial image, extracts the feature points in the initial image, and compares them with the feature points in the target image Points are matched to initialize the target area; then 2D affine transformation is performed on the initialized target area to obtain the candidate target area, and the expanded edge is mapped to the candidate target area according to the 2D affine transformation for matching, and the particle weights are updated. Particles are resampled to obtain a position estimate of the target. The invention also discloses a dynamic target real-time tracking system based on the improved geometric particle filter. The invention can effectively improve the real-time tracking performance of dynamic target vision.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and more specifically relates to a method and system for real-time tracking of dynamic targets based on improved geometric particle filtering. Background technique [0002] Visual object tracking is a research hotspot and difficulty in the field of computer vision, which integrates advanced achievements in image processing, object detection, filtering, automatic control, artificial intelligence and other technical fields. Visual object tracking has a wide range of application scenarios, and can be used in intelligent monitoring of public places, automatic driving and unmanned aerial vehicles to identify and track objects, motion analysis of objects, visual navigation and other fields. For the tracking of dynamic targets, due to the randomness and rapidity of the target's movement, this puts forward higher requirements for the robustness and real-time performance of visual target tracking....

AI Technical Summary

Problems solved by technology

However, the traditional geometric particle filter uses the method of pixel matching to calculate the weight of particles. When the target template occupies a large number of pixels and the number of particles is large, this method requires a large amount of calculation and cannot meet the real-time requirements. When the pixel values ​​of the object are similar, the accuracy of the pixel matching method is low, which will lead to a large geometric particle filter tracking error or even tracking failure.

Images