Alternative hough forests real time target tracking method based on GPU (graphics process unit)

Abstract

An alternative hough forests real time target tracking method based on a GPU (graphics process unit) is provided. The alternative hough forests real time target tracking method based on the GPU comprises: 1) extracting a training sample set according to a video sequence to be tested; 2) setting the number of random trees in the alternative hough forests and the maximum depth of the forests; 3) giving the training samples different weights; 4) initializing root nodes of each random tree; 5) building the alternative hough forests; 6) employing a splitting strategy when the nodes are splitting; 7) storing the alternative hough forests into a texture memory of the GPU; 8) manually determining an object region and an object center of a first frame image in a video, and setting a search radius; 9) detecting a follow-up frame through employing built alternative hough forests in the GPU, and obtaining a confidence map with respect to the object center position; 10) storing the confidence map into the texture memory of the GPU; 11) determining the object position in the present frame through employing the confidence map and combining the object region and the object center of the last frame; and 12) repeating the step 9 and the step 10 until completing the object tracking of all the frames in the video sequence.

Description

technical field [0001] The invention relates to image processing, in particular to a GPU-based alternate Hough forest real-time target tracking method that can be used in the fields of intelligent monitoring, target tracking and human-computer interaction. Background technique [0002] Visual object tracking is an important and complex research content in the field of computer vision. After years of research, it has become one of the research hotspots in the field of computer vision. Visual object tracking refers to locating a specific object in a video sequence, obtaining its motion parameters, and then estimating its pose. Target tracking has a very wide range of applications in many fields, including automatic target detection, target monitoring, target activity analysis and human-computer interaction, and is widely used in factories, schools, transportation, hospitals, banks and other places. The intelligent video surveillance system is based on target tracking, and aft...

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a GPU-based Alternating Hough Forest real-time target tracking method that can achieve stable and robust real-time target tracking for the problems of illumination changes, background interference, occlusion and low real-time performance in target tracking.

Images