|
OALib Journal期刊
ISSN: 2333-9721
费用：99美元
|
|
|
|
基于最优DAGSVM的服务机器人交互手势识别
DOI: 10.11834/jig.20090121
Keywords: 手势识别,小波矩,有向无环图支持向量机,人机交互
Abstract:
针对目前服务机器人手势交互方法在输入方式自然性和识别方法可靠性方面的不足，提出采用结合人脸和人手的姿态作为输入方式，实现了一个基于最优有向无环图支持向量机(DAGSVM)的手势识别系统。系统采用分步细化特征检测过程，即先粗检肤色，然后分别利用人眼Gabor特征和人手边缘小波矩特征检测脸和手部，可克服背景中的肤色干扰，并显著提高特征提取的可靠性；综合利用脸手区域不变矩和手的位置信息组成混合特征向量，采用优化拓扑排序策略组织多个两分类支持向量机(SVM)，构成最优DAGSVM多分类器，达到比普通DAGSVM更高的多分类准确率。实验验证了该方法的有效性和可靠性，并用于实现一种自然友好的人机交互方式。
| [1] | Waldherr S,Romero R,Thrun S,A gesture based interface for human-robot interaction,Autonomous Robots,2000(2).
|
| [2] | Stiefelhagen R,Fugen C,Gieselmann P,Natural human-robot interaction using speech,head pose and gestures,Sendai,Japan,2004.
|
| [3] | Kuno Y,Murashima T,Shimadat N,Understanding and learning of gestures through human-robot interaction,Takamatsu,Japan,2000.
|
| [4] | Hsu C W,Lin C J,A comparison of methods for multi-class Support Vector Machines,IEEE Transactions on Neural Networks,2002(2).
|
| [5] | 胡春华.马旭东.戴先中.钱？ 一种基于标准混合高斯模型的快速人脸检测方法 [J].中国图象图形学报A2007(3)
|
| [6] | Vapnik V,Chapelle O,Bounds on error expectation for support vector machines,Neural Computation,2000(9).
|
| [7] | Kortenkamp D,Huber E,Bonasso R P,Recognizing and interpreting gestures on a mobile robot,Portland,USA,1996.
|
| [8] | Richarz J,Martin C,Scheidig A,Estimating pointing gestures in monocular images for mobile robot instruction,Hatfield,UK,2006.
|
| [9] | Boehme H J,Braumann U D,Corradini A,Person localization and posture recognition for human-robot interaction,London:Springer-Verlag,1999.
|
| [10] | Platt J C,Cristianini N,Large margin DAGs for muhiclass classification,Advances in Neural Information Processing Systems,2000(3).
|
| [11] | Hsu R L,Mottaleb M A,Jain A K,Face detection in color images,IEEE Transactions on Pattern Analysis and Machine Intelligence,2002(5).
|
| [12] | Shen D G,Horace H S,Discriminative wavelet shape descriptors for recognition of 2-D patterns,Pattern Recognition,1999(2).
|
Full-Text
|
|
Contact Us
service@oalib.com QQ:3279437679 
WhatsApp +8615387084133
|
|
