单片机与嵌入式系统应用Issue(7)：10-13,4.
改进 mean shift 算法在目标跟踪中的研究
Research on Target Tracking Based on Improved mean-shift AIgorithm
周春云 1丁伟 2房伟 3谢林柏3
作者信息
- 1. 国家高性能计算中心扬州分中心，万方电子技术有限责任公司，扬州 225006
- 2. 扬州大学
- 3. 江南大学
- 折叠
摘要
Abstract
The improved mean-shift algorithm based on texture and color model is proposed,which can overcome the disadvantage of sin-gle target feature in the traditional target tracking.In the framework of traditional mean-shift algorithm,the system establishes the matc-hing feature from multiple perspectives,by fusing color model and texture model obtained from Contourlet transform.The proposed algo-rithm has been realized on an embedded platform.The experimental results show that the tracking effect is improved.关键词
目标跟踪/mean shift 算法/Contourlet 变换/嵌入式系统Key words
target tracking/mean-shift algorithm/Contourlet transform/embedded system分类
计算机与自动化引用本文复制引用
周春云,丁伟,房伟,谢林柏..改进 mean shift 算法在目标跟踪中的研究[J].单片机与嵌入式系统应用,2015,(7):10-13,4.
