郑州轻工业学院学报（自然科学版）2013，Vol.28Issue(3)：51-55,5.DOI:10.3969/j.issn.2095-476X.2013.03.012
基于几何矩的自适应手势跟踪算法研究
Research of self-adaption gesture tracking algorithm based on the geometric moment
摘要
Abstract
A new algorithm was proposed which based on the Mean-Shift algorithm.This algorithm can automatically select tracking gestures by combining skin color detection with seeking maximum connected domain;it using image geometric moment to alternate the Mean-Shift vector,and can also effectively track gesture target and achieve the size of tracking window changing along with the gestureg change of size,angle and shape.The experimental results showed that the algorithm can successfully track the dynamic and deformable gestures and capture its trajectory in real time.This algorithm processes about 27 frames persecond,and can meet the real-time requirements.关键词
手势跟踪/Mean-Shift算法/连通域/肤色检测/几何矩Key words
gesture tracking/Mean-Shift algorithm/connected domain/skin color detection/geometric moment分类
信息技术与安全科学引用本文复制引用
周荣强,全海燕,熊志伟,张天飞..基于几何矩的自适应手势跟踪算法研究[J].郑州轻工业学院学报（自然科学版）,2013,28(3):51-55,5.基金项目
云南省自然科学基金项目(2009ZC048M) （2009ZC048M）
昆明理工大学校人才培养基金项目(KKZ3201103022) （KKZ3201103022）
