中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
在线多线索动态手势识别

文献类型:期刊论文

作者赵莹; 迟冬祥; 胡静; 李菁辉
刊名合肥工业大学学报. 自然科学版
出版日期2013
卷号36期号:9页码:1071
其他题名Online dynamic hand gesture recognition with multiple cues
通讯作者赵莹, 上海电机学院电子信息学院, 上海 200240, 中国. ; 迟冬祥, 上海电机学院电子信息学院, 上海 200240, 中国. ; 胡静, 上海电机学院电子信息学院, 上海 200240, 中国. ; 李菁辉, 中国科学院上海光学精密机械研究所, 上海 201100, 中国.
中文摘要文章提出一种将运动和肤色特征相融合的多线索识别方法,利用差分运动检测降低复杂背景干扰,利用肤色在HSV空间中的聚类特性,采用多高斯模型,对每一个输入视频,将前50帧作为学习样本,在线学习出该视频的肤色分类器,解决了泛化性能低的问题;利用手势固有形状特征进行静态手势分割,采用基于仿射变换的块匹配法对动态手势进行跟踪。结果表明该算法能够较好地适应不同光照及复杂背景,同时满足实时要求。
英文摘要An online dynamic hand gesture recognition method with multiple cues based on the motion and the features of skin color is proposed. The disturbance caused by complex background is reduced by motion detection. According to the cluster characteristic of skin color in HSV space, and by using the multi-Gaussian model, the first 50 frames of each input video are used as the learning sample to construct the online skin classifier, which can solve the low generalization performance problem. The static hand recognition is completed with geometric feature. An affine model is adopted for the motion displacement estimation in hand tracking. The experimental results show that the proposed method is robust and real-time, and is able to adapt to different illumination and complex background.
收录类别EI
语种中文
版本出版稿
源URL[http://ir.siom.ac.cn/handle/181231/14881]  
专题上海光学精密机械研究所_信息光学与光电技术实验室
作者单位1.Zhao Ying, College of Electronic Information, Shanghai Dianji University, Shanghai 200240, China.
2.Chi Dongxiang, College of Electronic Information, Shanghai Dianji University, Shanghai 200240, China.
3.Hu Jing, College of Electronic Information, Shanghai Dianji University, Shanghai 200240, China.
4.Li Jinghui, Shanghai Institute of Optics and Fine Mechanics, Chinese Academy of Sciences, Shanghai 201100, China.
推荐引用方式
GB/T 7714
赵莹,迟冬祥,胡静,等. 在线多线索动态手势识别[J]. 合肥工业大学学报. 自然科学版,2013,36(9):1071.
APA 赵莹,迟冬祥,胡静,&李菁辉.(2013).在线多线索动态手势识别.合肥工业大学学报. 自然科学版,36(9),1071.
MLA 赵莹,et al."在线多线索动态手势识别".合肥工业大学学报. 自然科学版 36.9(2013):1071.

入库方式: OAI收割

来源:上海光学精密机械研究所

浏览0
下载0
收藏0
其他版本

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。