基于认知分析的手势跟踪方法研究  

COGNITIVE ANALYSIS BASED HAND GESTURE TRACKING METHOD STUDY

作　　者：冯志全[1,2] 郑艳伟[1,2] 徐涛[1,2] 蒋彦 

机构地区：[1]济南大学信息科学与工程学院,山东济南250022 [2]山东省网络环境智能计算技术重点实验室,山东济南250022

出　　处：《计算机应用与软件》2011年第11期257-260,共4页Computer Applications and Software

摘　　要：以降低粒子数目为目标,以分析操作者的认知模型为切入点,提出一种新的粒子滤波手势跟踪方法。首先,分析操作者在手势操作过程中的认知心理,建立人手运动的动态模型。其次,研究动态模型的基本特征,并给出一种描述方法。然后,提出一种基于状态变量微观结构的手势跟踪算法。结果表明,采用所介绍的算法,用少量粒子就可以得到比较精确的跟踪结果。Aiming at reducing particle numbers, starting from analyzing operators＇ cognitive model, a novel particle filtering hand gesture tracking approach is put forward. At first, operators＇ cognitive psychology during hand gesture operation processes is analyzed and a human hand motion dynamic model is established. Next, some fundamental characteristics of the dynamic model are studied and a description method is presented. Then, a status variable microstructure based hand gesture tracking algorithm is presented. Results show that, using only a small number of particles, comparatively accurate results can be obtained.

关 键 词：人手跟踪 粒子滤波 状态变量微观结构 降低粒子数目 认知心理 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]