摘要
为了提高实时性和准确性,提出一种改进的动态时间规整算法(Dynamic Time Warping-DTW),用于度量手势运动轨迹的相似性,实现了快速的精确动态手势识别.首先,通过Kinect2传感器实时地获取人体骨架的关节点坐标和手部的形状数据,然后构造矢量特征描述手的运动轨迹,运用动态时间规整方法进行模板匹配,并对特殊手势进行精确的二次分类,实现了基于轨迹匹配的快速动态手势识别.实验证明:该方法识别准确度高,实时性好,对光照强度和复杂背景干扰有很强的鲁棒性.
In order to improve the real-time performance and accuracy,an improved Dynamic Time Warping algorithm（ DTW） is proposed to measure the similarity of gesture trajectory and implement fast and accurate Dynamic gesture recognition. Firstly,the human body joint coordinates and the shape data of the hand are obtained based on Kinect2 sensor; secondly,we construct the vector features description gesture trajectory and use the DTW（ Dynamic Time Warping） method for template matching; finally,the second classification is performed for some special gestures to improve accuracy. Experiments proved that our method has high recognition accuracy and good real-time performance,and has a strong robustness to complex background and light intensity.
出处
《小型微型计算机系统》
CSCD
北大核心
2016年第7期1600-1603,共4页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(61374039
61403254)资助
沪江基金项目(C14002B1402/D1402
D15009)资助
关键词
动态手势识别
模板匹配
DTW
分类
dynamic gesture recognition
template matching
DTW algorithm
classification
