Affiliation of Author(s):计算机科学与技术学院/人工智能学院/软件学院
Journal:Lect. Notes Comput. Sci.
Abstract:Color feature is mainly adopted in the traditional particle filter method when tracking the target. In view of the problem of failing to track the target caused by background similarity and occlusion, an improved particle filter tracking algorithm based on color histogram and convolution network is proposed, which makes full use of the color feature and convolution feature of the target. Experiments show that compared with the traditional tracking algorithm based on particle filter, the proposed algorithm has a good ability to adapt to changes in the environment around the target. © Springer International Publishing AG, part of Springer Nature 2018.
ISSN No.:0302-9743
Translation or Not:no
Date of Publication:2018-01-01
Co-author:Gao, Shasha,xq
Correspondence Author:zl
Date of Publication:2018-01-01