基于表观特征分析的手势识别及其应用

针对复杂背景下的手势识别容易受到环境干扰造成的识别困难问题,通过分析手势的表观特征,提出并实现了一种可用于自然人机交互的手势识别算法。该算法基于Kinect深度图像实现手势区域分割,然后提取手势手指弧度、指间弧度、手指数目等具有旋转缩放不变性的表观特征,运用最小距离法实现快速分类。并将该算法成功运用于实验室三指灵巧手平台,达到了理想的控制效果。实验表明该算法具有良好的鲁棒性,针对九种常用手势,平均识别率达到94.3%。

Hand gesture recognition based on analysis of appearance features and its application

Hand gesture recognition in complex background is susceptible to environmental interference, thus leading to recognition difficulty. According to this recognition problem, by analyzing the appearance features of hand gesture, a hand gesture recognition algorithm for natural human computer interaction is proposed and implemented. By using depth images which obtain from Kinect, we extract features like gesture finger radians, radians between fingers and the number of fingers, and properly utilize minimum distance algorithm for achieving fast and efficient classification. Experimental results show that the algorithm is robust and real-time with an average recognition rate of 94.3% for nine frequently-used gestures.