基于表观建模的中国手语识别技术研究
|
摘要
研究中国手语的识别技术,就是为了把聋人使用的手语通过计算机自动、高效地完成机器翻译,从而实现手语与自然语音的无障碍交流。这种方式便于聋人融入社会,有利于他们与周围环境的交流,为其提供更好的服务。同时,手语识别在其他应用领域,也有着深远的研究意义。基于计算机视觉的手语识别作为一种自然、直观的交互方式,无需附加的物理设备作为输入,在人机交互过程研究中占很大比重,能够广泛地应用于多学科领域。作为一个非常有意义的研究课题,手语识别的研究,不仅有助于改善、提高聋人的生活、学习和工作环境,也能够提高计算机对人类自然语言的理解水平,发展成为一种能够付诸应用的最自然的人机交互方式。
本文从自然交互方式的角度出发,研究了基于机器视觉的手语手势跟踪、手势分割提取、手语表观建模、SVM核函数构建算法及中国手指语字母的识别。具体研究工作包括以下几个方面：
(1)根据Kinect同步拍摄深度视频信息的特点,使用手语视频中的深度图像信息对CamShift加以改进,提出了一种在复杂场景下跟踪能力更强、抗干扰性能更好的,适于手语识别应用的DI_CamShift (Depth Image CamShift)算法。采用基于深度图像信息的跟踪算法在手语视频中确定手势区域,对手语手势进行定位跟踪。在手势提取方面,通过计算手势深度图像确定手势的主轴方向,提出了一种基于深度图像信息的椭圆边界肤色建模方法。
(2)在复杂背景下的手势提取过程中,结合椭圆边界肤色模型,得出新的基于深度积分图像的二维OTSU算法,并将积分图和粒子群结合用于二维OTSU算法,提出了基于深度积分图和粒子群优化的OTSU算法提取手势图像。
(3)在手语表观建模时,为了使其各种视觉特征均可以被描述的更为准确,同时提取了SURF特征,Gabor小波纹理特征和颜色直方图特征,作为手势表观特征的完备特征集。在解决手势图像局部特征点个数不同的问题时,本文提出使用BoW (Bag of Words)方法,把提取出的手语完备特征集量化生成手语视觉单词(Sign Language Visual Word)。对提取到的手势特征,采用K-Means聚类算法,生成手语SURF词包、Gabor词包和颜色直方图词包,最后把所有词包经过CCA融合后的结果作为手语的特征。
(4)研究了SVM及核函数理论,通过提取相同的手势特征对单核SVM分类器进行训练,比对研究不同核函数SVM在手语识别中的效果。提出构建一种新的适于手语识别的核函数H_Kernel,并证明了H_Kernel满足Mercer条件,可以作为用于手语识别的SVM核函数。鉴于手语BoW模型并未考虑语义信息,所以本文提出构建基于H Kernel和BoW语义核的混合核函数SVM进行训练学习及手语识别。
Sign language recognition is the method which uses the computer to translate the sign language into text or voice. It can realize the communication between the natural language and the sign language. It provides good service for the deaf, to help the deaf communicate to normal people and enables them feel comfortable in the society. At the same time, the sign language recognition can also be used in other application fields, and has a significant importance. Sign language recognition, which is based on computer vision, is a natural and intuitive interaction. Without additional physical device, it has a great influence on the human-computer interaction research and can be widely used in many fields. As a very meaningful research subject, sign language recognition research not only help to improve the living, learning and working environment for the deaf, but also helpful to the computer's understanding level of human natural language and develops in the most natural way of human-computer interaction.
This paper studies sign language gesture tracking, gesture segmentation, gesture modeling based on appearance, constructing method of SVM kernel function, and Chinese finger alphabet recognition. Specific research work includes the following aspects:
(1) According to the characteristics of Kinect depth information video, the CamShift has been improved, and a new tracking algorithm is developed, which is the DI_CamShift (the Depth Image CamShift) algorithm. Under the complex scene, its tracking ability is stronger, more anti-jamming and suitable for sign language recognition. After the DI_CamShift algorithm tracking in the sign language video, gesture is located. The principal axis direction is determined by calculation of depth image, and a new modeling method based on depth image information is developed.
(2) Under the complex background, gestures extraction combines with the elliptic boundary model, and produces a new2D OTSU algorithm based on integral depth image. Then both of the integral figure and the particle swarm is used in2D OTSU algorithm to develop an optimization algorithm.
(3) While the sign language apparent modeling, complete feature set of the gesture is constructed in order to describe the various visual features accurately, which includes the SIFT feature, Gabor characteristics, SURF feature and color histogram. To solve the problem of feature points number, this paper uses the BoW (Bag of Words) method. The sign language feature set is used to generate sign language visual word. All of the gesture features are clustered by K-Means algorithm, and generated the SURF BoW, Gabor BoW and color histogram BoW. All of the BoWs are fusioned together by CCA.
(4) While study of SVM theory and kernel function, this paper trains the SVM classifier by extracting the same gestures feature. A new kernel named H_Kernel, is developed and be proved as satisfy to Mercer condition after compared the effect of SVM with different kernel functions in sign language recognition. It's also suitable for sign language recognition. As BoW model does not contain the sign language semantic information, a method which based on H_Kernel and BoW semantic kernel function is proposed for SVM learning and sign language recognition.
本文从自然交互方式的角度出发,研究了基于机器视觉的手语手势跟踪、手势分割提取、手语表观建模、SVM核函数构建算法及中国手指语字母的识别。具体研究工作包括以下几个方面：
(1)根据Kinect同步拍摄深度视频信息的特点,使用手语视频中的深度图像信息对CamShift加以改进,提出了一种在复杂场景下跟踪能力更强、抗干扰性能更好的,适于手语识别应用的DI_CamShift (Depth Image CamShift)算法。采用基于深度图像信息的跟踪算法在手语视频中确定手势区域,对手语手势进行定位跟踪。在手势提取方面,通过计算手势深度图像确定手势的主轴方向,提出了一种基于深度图像信息的椭圆边界肤色建模方法。
(2)在复杂背景下的手势提取过程中,结合椭圆边界肤色模型,得出新的基于深度积分图像的二维OTSU算法,并将积分图和粒子群结合用于二维OTSU算法,提出了基于深度积分图和粒子群优化的OTSU算法提取手势图像。
(3)在手语表观建模时,为了使其各种视觉特征均可以被描述的更为准确,同时提取了SURF特征,Gabor小波纹理特征和颜色直方图特征,作为手势表观特征的完备特征集。在解决手势图像局部特征点个数不同的问题时,本文提出使用BoW (Bag of Words)方法,把提取出的手语完备特征集量化生成手语视觉单词(Sign Language Visual Word)。对提取到的手势特征,采用K-Means聚类算法,生成手语SURF词包、Gabor词包和颜色直方图词包,最后把所有词包经过CCA融合后的结果作为手语的特征。
(4)研究了SVM及核函数理论,通过提取相同的手势特征对单核SVM分类器进行训练,比对研究不同核函数SVM在手语识别中的效果。提出构建一种新的适于手语识别的核函数H_Kernel,并证明了H_Kernel满足Mercer条件,可以作为用于手语识别的SVM核函数。鉴于手语BoW模型并未考虑语义信息,所以本文提出构建基于H Kernel和BoW语义核的混合核函数SVM进行训练学习及手语识别。
Sign language recognition is the method which uses the computer to translate the sign language into text or voice. It can realize the communication between the natural language and the sign language. It provides good service for the deaf, to help the deaf communicate to normal people and enables them feel comfortable in the society. At the same time, the sign language recognition can also be used in other application fields, and has a significant importance. Sign language recognition, which is based on computer vision, is a natural and intuitive interaction. Without additional physical device, it has a great influence on the human-computer interaction research and can be widely used in many fields. As a very meaningful research subject, sign language recognition research not only help to improve the living, learning and working environment for the deaf, but also helpful to the computer's understanding level of human natural language and develops in the most natural way of human-computer interaction.
This paper studies sign language gesture tracking, gesture segmentation, gesture modeling based on appearance, constructing method of SVM kernel function, and Chinese finger alphabet recognition. Specific research work includes the following aspects:
(1) According to the characteristics of Kinect depth information video, the CamShift has been improved, and a new tracking algorithm is developed, which is the DI_CamShift (the Depth Image CamShift) algorithm. Under the complex scene, its tracking ability is stronger, more anti-jamming and suitable for sign language recognition. After the DI_CamShift algorithm tracking in the sign language video, gesture is located. The principal axis direction is determined by calculation of depth image, and a new modeling method based on depth image information is developed.
(2) Under the complex background, gestures extraction combines with the elliptic boundary model, and produces a new2D OTSU algorithm based on integral depth image. Then both of the integral figure and the particle swarm is used in2D OTSU algorithm to develop an optimization algorithm.
(3) While the sign language apparent modeling, complete feature set of the gesture is constructed in order to describe the various visual features accurately, which includes the SIFT feature, Gabor characteristics, SURF feature and color histogram. To solve the problem of feature points number, this paper uses the BoW (Bag of Words) method. The sign language feature set is used to generate sign language visual word. All of the gesture features are clustered by K-Means algorithm, and generated the SURF BoW, Gabor BoW and color histogram BoW. All of the BoWs are fusioned together by CCA.
(4) While study of SVM theory and kernel function, this paper trains the SVM classifier by extracting the same gestures feature. A new kernel named H_Kernel, is developed and be proved as satisfy to Mercer condition after compared the effect of SVM with different kernel functions in sign language recognition. It's also suitable for sign language recognition. As BoW model does not contain the sign language semantic information, a method which based on H_Kernel and BoW semantic kernel function is proposed for SVM learning and sign language recognition.
引文
[1]Juan Pablo Wachs, Mathias Kolsch, Helman Stern, et al.Vision-Based Hand-Gesture Applications[J]. Communications of the ACM, February 2011, Vol.54 No.2:60-71
[2]马继勇.手语理解的统计模型研究[D].北京：中国科学院计算技术研究所,2001
[3]郑重雨.基于运动特征的手语动画关键帧调整和插值方法的研究[D].北京：北京工业大学,2009
[4]中国聋人协会.中国手语(修订版)[M].北京：北京华夏出版社,2003.4
[5]李连昌.复杂场景下的视觉目标跟踪方法研究[D].济南：山东大学,2010
[6]段宏湘.序列图像中动态手势跟踪与建模方法研究[D].兰州：兰州理工大学,2011
[7]张毅,张姣,罗元.基于手势跟踪的智能轮椅控制系统[J].重庆邮电大学学报,2011.6：73-76
[8]翁汉良,战荫伟.基于视觉的多特征手势识别[J].计算机工程与科学,2012.2：123-127
[9]王超,崔更申,黄延辉等.线上线下结合的实时手势跟踪系统[J].桂林电子科技大学学报,2012.2：95-99
[10]王西颖,张习文,戴国忠.一种面向实时交互的变形手势跟踪方法[J].软件学报,2007.10：2423-2433
[11]Wang R, Popovic J. Real-Time Hand-Tracking with a Color Glove[J]. ACM Trans on Graphics,2009,28(3):1-8
[12]冯志全,杨波,李毅等.以时间优化为目标的粒子滤波手势跟踪方法研究[J].电子学报,2009.9：1989-1995
[13]常亚南.基于HMM的动态手势识别[D].华南理工大学,2012
[14]王西颖,戴国忠,张习文等.基于HMM-FNN模型的复杂动态手势识别[J].软件学报,2008.9：2302-2312
[15]彭娟春,顾立忠,苏剑波.基于Camshift和Kalman滤波的仿人机器人手势 跟踪[J].上海交通大学学报,2006.7：1161-1165
[16]张秋余,胡建强,张墨逸.基于区域生长的Mean shift动态变形手势跟踪算法[J].模式识别与人工智能,2010.4：580-585
[17]甘明刚,陈杰,王亚楠等.基于Mean Shift算法和NMI特征的目标跟踪算法研究[J].自动化学报,2010.9：1332-1336
[18]王鑫,唐振民.一种改进的基于Camshift的粒子滤波实时目标跟踪算法[J].中国图象图形学报,2010.10：1507-1514
[19]Zhou Ren, Junsong Yuan, Zhengyou Zhang. Robust Hand Gesture Recognition Based on Finger-Earth Mover's Distance with a Commodity Depth Camera [A]. MM'11[C], Scottsdale, Arizona, USA, November 28-December 1,2011:1093-1096
[20]雷海龙,杨胜,刘红新.融合图像轮廓矩和傅立叶描述子的手势识别[J].微电子学与计算机,2011.4：550-558
[21]张海燕,王博亮.基于改进的Snake模型的眼细胞彩色图像边缘检测算法研究[J].厦门大学学报(自然科学版),2005,03：76-80
[22]曹听燕,赵继印,李敏.基于肤色和运动检测技术的单目视觉手势分割[J].湖南大学学报(自然科学版),2011.1：78-83
[23]杨波,宋晓娜,冯志全等.复杂背景下基于空间分布特征的手势识别算法[J].计算机辅助设计与图形学学报,2010.10：1841-1848
[24]Abbas Cheddad, Joan Condel. A skin tone detection algorithm for an adaptive approach to steganography[J]. Signal Processing,2009,89(12):2456-2478
[25]高建坡,王煜坚,杨浩等.一种基于KL变换的椭圆模型肤色检测方法[J].电子与信息学报,2007(07)：1739-1743
[26]王鼎,沈辉,娄海涛.一种基于H-CrCb颜色空间的肤色检测算法研究[J].计算机科学,2012(39)：183-187
[27]张登康,郭太良,姚剑敏等.改进肤色识别参量和轮廓特征相结合的指尖检测[J].光电子技术,2012.3：185-189
[28]Argyros AA, Lourakis MIA. Real time tracking of multiple skin colored objects with a possibly moving camera [A]. In Proc. ECCV[C],2004.3:368-379
[29]包加桐,宋爱国,郭晏.基于SURF特征跟踪的动态手势识别算法[J].机器人, 2011.4：482489
[30]陈意,杨平,陈旭光.一种基于加速度特征提取的手势识别方法[J].传感技术学报,2012.8：1073-1078
[31]梁振.基于辐射投影变换和极半径不变矩的静态手势识别研究[D].山东科技大学,2009
[32]张墨逸.基于表观的二维手势识别方法研究[D].兰州：兰州理工大学,2010
[33]文学志,方巍,郑钰辉.一种基于类Haar特征和改进AdaBoost分类器的车辆识别算法[J].电子学报,Vol.39, No.5,2011.5:1121-1126
[34]王凯,于鸿洋,张萍.基于AdaBoost算法和光流匹配的实时手势识别[J].微电子学与计算机, Vol.29, No.4,2012.4:138-141
[35]Paul Doliotis, Alexandra Stefan, Christopher McMurrough, et al. Comparing Gesture Recognition Accuracy Using Color and Depth Information [A]. PETRA 2011[C], Crete, Greece,2011:1-7
[36]覃文军,吴成东,赵姝颖等.基于傅立叶描述子-BP神经网络的手势识别方法[J].东北大学学报(自然科学版),Vol.30, No.9,2009.9:1232-1235
[37]谭昶,肖南峰.基于改进RCE和RBF神经网络的静态手势识别[J].计算机工程与应用,2011.47(7)：172-176
[38]李文生,解梅,邓春健.一种快速的动态手势学习和识别方法[J].南京大学学报(自然科学版),2012.4：421-428
[39]Yun Liu, Yanmin Yin, Shujun Zhang. Hand Gesture Recognition Based on HU Moments in Interaction of Virtual Reality [A].4th International Conference on Intelligent Human-Machine Systems and Cybernetics[C],2012:145-148
[40]阳平,陈香,李云等.一种基于融合多传感器信息的手语手势识别方法[J].航天医学与医学工程,2012.4：276-281
[41]王德鑫,石崇林,张茂军.基于Petri网和BPNN的多重触控手势识别[J].模式识别与人工智能,2010.3：408-413
[42]孙荣斌.基于粒子滤波的手势识别技术研究[D].南京：南京理工大学,2009
[43]王文会,陈香,阳平等.基于多传感器信息检测和融合的中国手语识别研究[J].中国生物医学工程学报,2010.5：665-671
[44]ULRYCHJ, KOPECKY M. Visual similarity in sign language [A]. Proc. of the 24th International Conference on Data Engineering[C],2008:346-353
[45]Priyal S P, Bora P K. A study on static hand gesture recognition using moments[A]. Proc of IEEE SPCOM'10[C], Bangalore IEEE,2010:1-5
[46]P.Dreuw, J.Forster, H.Ney. Tracking Benchmark Databases for Video-Based Sign Language Recognition[A]. In ECCV International Workshop on Sign, Gesture and Activity(SGA)[C]. Crete, Greece, September 2010:335-341
[47]P.Dreuw, D.Rybach, T.Deselaers, et al. Speech Recognition Techniques for a Sign Language Recognition System[R]. In Interspeech, Antwerp, Belgium, August 2007. ISCA best student paper award Interspeech,2007
[48]吴江琴,高文,陈熙霖等.基于ANN/HMM的中国手语识别系统[J].计算机工程与应用,1999.9：1-4
[49]Wen Gao, Ji-Yong Ma, Jiang-Qin Wu, et al. Large vocabulary sign language recognition based on HMM/ANN/DP[J]. International Journal of Pattern Recogniton and Artificial Intelligence,2000,14(5):587-602
[50]杨筱林,姚鸿勋.基于多尺度形状描述子的手势识别[J].计算机工程与应用,2004,40(32)：76-78
[51]张良国,高文,陈熙霖等.面向中等词汇量的中国手语视觉识别系统[J].计算机研究与发展,2006,43(3)：476-482
[52]姜峰,高文,姚鸿勋等.非特定人手语识别问题中的合成数据驱动方法[J].计算机研究与发展,2007,44(5)：873-881
[53]Zhu Yuanxin, XuGuangyou, Huang Yu. Appearance-based dynamic hand gesture recognition from image sequences with complex background[J]. Journal of Software, 2000,11(1):54-61
[54]ZHANG Shi-lin, ZHANG Bo. Using HMM to sign language video retrieval[A]. Proc of the 2nd International Conference on Computational Intelligence and Natural Computing[C].2010:55-59
[55]Fels. Glove talk II:Mapping hand gestures to speech using neural networks-an approach to building adaptive interfaces[D]. Computer Science Department, University of Torino,1994
[56]T.Takahashi, F.K.Shino. Hand gesture coding based on experiments using a handgesture interface device[J]. SIGCHI Bulletin,1991,23(2):67-73
[57]Liang R H, Ouhyoung M, A real-time continuous gesture recognition system for sign language[A]. Proc of 3rd International Conference on Automatic Face and Gesture Recognition[C]. Nara, Japan,1998:558-565
[58]Y Cui, D L Swets, J JWeng. Learning-based hand sign language using SHOSLIF-M[A]. Proc of ICCV'96[C].1996:621-636
[59]Triesch J, MalsburgC.Robust Classification of Hand Postures Against Complex Background[A]. International Conference on Automatic Face and Gesture Recognition[C].1996:170-175
[60]Quek F, Zhao M. Inductive Learning in Hand Pose Recognition[J], IEEE Automatic Face and Gesture Recognition,1996:78-83
[61]Davis, ShahParallel. Hidden Markov Models for American Sign Language Recognition[A]. Proceedings of the International Conference on Computer Vision[C]. Kerkyra, Greece, September,1999:22-25
[62]Starner T, Pentland A. Real-time American sign language recognition from video using hidden Markov models[R]. Media Lab, MIT:Technical Report TR375,1996
[63]Christian Vogler, Dimitris Metaxas. Adapting hidden Markov models for ASL recognition by Using three-dimensional computer vision methods[A]. Proceedings of the IEEE International Conference On Systems, Man and Cybernetics[C].Orlando, FL, 1997:156-161
[64]ChristianVogler, Dimitris Metaxas. ASL recognition based on a coupling between HMMS and 3D motion analysis[A]. Proceedings of the IEEE International Conference on Computer Vision[C].Mumbai, India,1998:365-371
[65]ChristianVogler, Dimitris Metaxas. Toward scalability in ASL Recognition: Breaking Down Signs into Phonemes[A]. Proceedings of Gesture Workshop[C]. Gif-sur-Yvette, France,1999:400-404
[66]ChriatianVogler, DimitrisMetaxa. Parallel hidden Markov models for American Sign Language recognition[A]. Proceedings of the International Conference on Computer Vision[C].Kerkyra, Greece,1999:224-228
[67]J.W. Deng, H. T. Tsui. A two-step approach based on HMM for the recognition of ASL[A]. The 5th Asian Conf. Computer Vision[C]. Melbourne, Australia, 2002:363-369
[68]CHEN Qing, GEORGANAS N D, PETRIU E M. Real-time vision-based hand gesture recognition using Haar-like features[A]. Proc. of IEEE Instrumentation and Measurement Technology Conference[C]. IEEE Press,2007:1-6
[69]SILANON K, SUVONVORN N. Hand motion analysis for Thai alphabet recognition using HMM[J]. International Journal of Informationand Electronics Engineering,2011,1(1):65-71
[70]Ong.Eng-Jon, Cooper.Helen, Pugeault.Nicolas, et al. Sign Language Recognition using Sequential Pattern Trees[A]. The IEEE Conference on Computer Vision and Pattern Recognition(CVPR) [C]. Providence, Rhode Island,16-21 June, 2012:2200-2207
[71]NIEBLES J, WANG H, LI F F. Unsupervised learning of human action categories using spatial temporal words [J]. International Journal of Computer Vision, 2008,79(3):299-318
[72]DARDAS N H, PETRIU E M. Hand gesture detection and recognition using principal component analysis[A]. Proceedings of International Conference on Computational Intelligence for Measurement Systems and Applications (CMS A) [C]. Ottawa:IEEE,2011:1-6
[73]黄康泉,陈壁金,郑博等.Kinect在视频会议系统中的应用[J].广西大学学报：自然科学版,2011,36：308-310
[74]Zheng Xiao, Fu Mengyin, Yang Yi et al.3D Human Postures Recognition Using Kinect[A].4th International Conference on Intelligent Human-Machine Systems and Cybernetics (IHMSC2012) [C]. Jiang Xi, Nan Chang, P.R.China, 2012.8.26-27:1796-1801
[75]RAHEJA J L, CHAUDHARV A, SIGNAL K. Tracking of fingertips and centers of palm using Kinect[A]. Proc. of the 3rd International Conference on Computational Intelligence, Modeling and Simulation[C]. IEEE Press,2011:248-252
[76]齐苏敏.序列图像中手势跟踪与识别技术的研究[D].苏州：苏州大学,2008
[77]Lipton A.J, Fujiyoshi H., Patti R.S..Moving target classification and tracking from real-time video[A]. Proceedings of IEEE Workshop on Applications of ComputerVision[C].Princeton, NJ, USA,1998:3642-3647
[78]刘亚辉.面向智能空间的多视角视觉系统关键技术研究[D].北京：北京邮电大学,2011
[79]Jin Li, JunhongXu,Wang Cong, et al. Research on real-time detection of moving target using gradient optical flow[A]. Proceedings of IEEE International Conference on Mechatronics and Automation[C].Niagara Fails, CanMa,2005,4:1796-1801
[80]McKenna S., J. JabriS, Duric Z, et al. Tracking groups of people[J]. Computer Vision and Image Understanding.2000,80(1):42-56
[81]Yamazawa K, Yokoya N. Detecting moving objects from omnidirectional dynamic images based on adaptive background subtraction[A]. Proceedings of International Conference on Image Processing[C].2003,3:953-956
[82]Stauffer C, Grimson W E L. Adaptive background mixture models for realtime tracking[A]. Computer Vision and Pattern Recognition[C], IEEE Computer Society Conference,1999,2:8-14
[83]R.Collin. A system for video surveillance and monitoring VSAM final report[R].Carnegie Mellon University,Technical Report:CMU-IR-TR-00-12,2000
[84]Paul Doliotis, Alexandra Stefan, Christopher McMurrough, et al. Comparing Gesture Recognition Accuracy Using Color and Depth Information[A]. PETRA 2011 [C]. Crete, Greece:20-23
[85]P. Perez, J. Vemaak and ABlake.Data fusion for visual tracking with particle[A]. Proceedings of the IEEE[C].2004,92(3):495-513
[86]R.T.Collies, Y.X.Liu and M.Leordeanu. Online selection of discriminative tracking features[A].IEEE Transactions on Pattern Analysis and Machine Intelligence[C].2005,27(10):1631-1643
[87]E.Wan and R.Van der Mexwe. The unscented Kalman Filter[A].S.Hayhin(Eels.), Kalman Filtering and Neural Networks[C].John Wiley & Sons Inc,2001:221-280
[88]K.Nummiaro, E.Koller-Meier, L.Van GOOL. An adaptive color-based particle filter[J]. Image and Vision Computing.2003,21(1):99-110
[89]任仙怡,廖云涛,张桂林等.一种新的相关跟踪方法研究[J].中国图形图像学报.2002.7(6)：553-557
[90]Hager GD, Dewan M, Stewart C V. Multiple Kernel Tracking with SSD[A]. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition[C].2004:790-797
[91]C.Yang, R.Duraiswami, L.Davis. Efficient mean-shift tracking via a new similarity measure[A]. Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition[C].San Diego, USA,2005:176-183
[92]Zhaowen Wang, Xiaokang Yang, Yi Xu, et al. CamShift guided particle filter for visual tracking[J]. Pattern Recognition Letters.30(2009):407-413
[93]J.Kovacevic, S.Juric-Kavelj, I.Petrovic, et al. An Improved CamShift Algorithm Using Stereo Vision For Object Tracking[A]. MIPRO 2011[C]. Opatija, Croatia, May 23-27,2011:707-710
[94]Kinect[EB/OL]. http://en.wikipedia.org/wiki/Kinect,2013.3
[95]Windows Kinect SKD[EB/OL]. http://kinectforwindows.org,2013.3
[96]Prime Sense OpenNI[EB/OL]. http://www.primesense.com,2013.3
[97]罗元,谢或,张毅.基于Kinect传感器的智能轮椅手势控制系统的设计与实现[J].机器人ROBOT. Vol.34, No.l, Jan.,2012,34(1):110-113
[98]文志强.基于均值偏移算法的运动目标跟踪技术的研究[D].长沙：中南大学,2008
[99]王帅.一种基于MeanShift算法的目标跟踪算法的研究[D].济南：山东大学,2011
[100]Otsu N. A Threshold Selection Method from Gray-level Histograms[J]. IEEE Trans on SMC.2009(1):62-69
[101]CARREIRA PERPINAN M A. Acceleration strategies for Gaussian Mean Shift image segmentation[A]. Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition[C]. New York,2006:543-549
[102]赵雪.基于自适应颜色特征的CamShift改进算法的研究[D].大连：大连海事大学,2011
[103]胡铟,杨静宇.基于分块颜色直方图MeanShift跟踪算法[J].系统仿真学报.2009,10：2936-2939
[104]何革.基于MeanShift算法的视频目标追踪研究[D].重庆：重庆大学,2010
[105]于永健,王向阳,吴俊峰.基于颜色复杂度的加权颜色直方图图像检索算法[J].小型微型计算机系统.2009,Vol-30.No.3:507-511
[106]Geetha M, Manjusha U C. A Vision Based Recognition of Indian Sign Language Alphabets and Numerals Using B-Spline Approximation[J]. International Journal on Computer Science and Engineering, Vol.4 No.03 March 2012:578-582
[107]RAHEJA J.L, CHAUDHARV A, SIGNAL K. Tracking of Fingertips and Centres of Palm using Kinect[A]. CIMSIM[C], Langkawi, IEEE,2011:248-252
[108]OpenKinect Organization. Imaging Information for Kinect[EB/OL]. http://openkinect.org/wiki/Imaging_Information,2013
[109]陈锻生,刘政凯.肤色检测技术综述[J].计算机学报,2006,29(2)：194-207
[110]曹听燕,赵继印,李敏.基于肤色和运动检测技术的单目视觉手势分割[J].湖南大学学报(自然科学版).Vlo.38, No.1,2011.1:78-83
[111]吴海滨,周后伟,张铁译等.基于OTSU的动态结合全局阈值的图像分割[J].大气与环境光学学报.Vol.7, No.6,2012.11:463-465
[112]范九伦,赵凤.灰度图像的二维Otsu曲线阈值分割法[J].电子学报.2007,4：751-755
[113]刘春生,常发亮,陈振学等.改进的高斯肤色模型及其在人脸检测中的应用[J].仪器仪表学报.2012,33(5)：1117-1121
[114]朱志亮,刘富国,陶向阳.基于积分图和粒子群优化的肤色分割[J].计算机工程与应用.2013.1：789-792
[115]郎咸朋,朱枫,都颖明等.基于积分图像的快速二维Otsu算法[J].仪器仪表学报,2009,30(1)：39-43
[116]Jang H, Do J H, Jung J, et al. View-invariant hand-posture recognition for soft-remo control system[A]. IEEE International Conference on Intelligent Robots and Systems[C], Sendai, Japan:IEEE,2004,1:295-300
[117]J.Zaletelj, J.Perhavc, J.F. Tasic. Vision-based Human-computer Interface Using Hand Gestures[A].2007 International Workshop on Image Analysis for Multimedia Interactive Services[C].Santorini,Greece,2007:1-4
[118]J.F.Xu, S.Q.Jun, Y.W.Shi, et al. Implementation of Digital Chime-bell Interaction System Driven by Hand Gesture[A]. The 2008 Interactional Conference on Pervasive Computing and Applications[C].Alexandria, Egypt,2008:264-268
[119]朱建伟.基于多摄像头的手势识别研究与实现[D].广州：中山大学,2010.6
[120]王骐,陈熙霖,王春立等.一种可处理数据缺失的视角无关手语识别方法[J].计算机学报,2009.5：954-960
[121]华斌,夏利娜.基于中值滤波和Hu矩向量的手语识别[J].计算机工程与设计.2011.2：615-618
[122]倪训博,赵德斌,高文.非特定人手语数据生成及其有效性检测[J].软件学报,2010.5：151-163
[123]汪道寅.基于SIFT图像配准算法的研究[D].合肥：中国科学技术大学,2011.5
[124]Morel J M, Yu G S. ASIFT:A New Framework for Fully Affine Invariant Image Comparison[J]. Society for Industrial and Applied Mathematics Journal on Image Sciences,2009,2(2):438-469
[125]Song R, Szymanski J. Well-distributed SIFT features[A]. Electronics Letters[C], 2009,45(6):308-310
[126]Zhou H Y, Yuan Y, Shi C M. Object tracking using SIFT features and mean shift[J]. Computer Vision and Image Understanding,2009,113(3):345-352
[127]蔺海峰,马宇峰,宋涛.基于SIFT特征目标跟踪算法研究[J].自动化学报,2010.8：1204-1208
[128]Juan L,Gwun O. A Comparison of SIFT, PCA-SIFT and SURF[J]. International Journal of Image Processing,2009,3(4):143-152
[129]陈小丹,杜宇人,高秀斌.一种基于SURF的图像特征点快速匹配算法[J]. 扬州大学学报(自然科学版),2012.4：15-18
[130]Y Bastanlar,A Temizel,Y Yardimci.Improved SIFT matching for image pairs with scale difference[J]. Electronics Letters,2010.6:346-348
[131]Luo Juan.A comparison of sift, pea-sift and surf[J]. International Journal of Image Processing,2009.12:143-152
[132]范新南,丁朋华,刘俊定等.融合灰度和SURF特征的红外目标跟踪[J].中国图象图形学报,2012.11：1376-1383
[133]卢振坤,玉振明.基于自适应局部Log-Gabor能量的多聚焦图像融合[J].电子学报,2013.1：35-39
[134]LI X, He M, Roux M. Multifocus image fusion based on redundant wavelet transform[J]. IET Image Processing,2010:283-293
[135]吴霁轩.基于二维Gabor特征和多特征融合的人脸识别研究[D].长春：吉林大学,2012
[136]耿佳佳,康晓东,王昊等.联合Gabor和分形的医学影像图像局部特征描述[J].计算机科学,2012,S3：97-101
[137]周健,王华彬,陶亮.基于离散傅里叶变换和块时间递归并行格型结构的离散Gabor分析窗求解[J].电子学报,2012.9：1839-1843
[138]王向阳,杨红颖,郑宏亮等.基于视觉权值的分块颜色直方图图像检索算法[J].自动化学报,2010.10：1489-1492
[139]陈海林,吴秀清.基于双空间金字塔匹配核的图像目标分类[J].中国科学技术大学学报,2010(03)：313-320
[140]贾世杰,孔祥维.一种新的直方图核函数及在图像分类中的应用[J].电子与信息学报,2011.7：1738-1742
[141]王宇石；高文.用基于视觉单词上下文的核函数对图像分类[J].中国图象图形学报,2010.4：607-616
[142]刘扬闻,霍宏,方涛.词包模型中视觉单词歧义性分析[J].计算机工程,Vol.37, No.19,2011.10:164-170
[143]张秋余,王道东,张墨逸等.基于特征包支持向量机的手势识别[J].计算机应用,2012,32(12)：3392-3396
[144]陆福相.基于内容的场景和物体类别识别研究[D].上海：上海交通大学,2011
[145]李新德,金晓彬,张秀龙.一种基于BoW物体识别模型的视觉导航方法[J].东南大学学报(自然科学版).2012.3：3-8
[146]武京相.融合全局和局部特征的医学图像分类[D].成都：电子科技大学,2010
[147]王科俊,邹国锋,阎涛.融合全局与局部特征的单样本人脸识别[J].中南大学学报(自然科学版),2011.9：1-7
[148]庄凌,庄越挺,吴江琴等.一种基于稀疏典型性相关分析的图像检索方法[J].软件学报,2012.5：89-94
[149]HAI Xianwang.Local two-dimensional canonical correlation analysis[J]. Signal Processing.2010.7:921-924
[150]刘瑞瑞.人脸特征提取及融合方法研究[D].南京：东南大学,2010
[151]梁锦锦.支持向量机和支持向量域描述的若干问题研究[D].西安：西安电子科技大学,2009
[152]彭新俊.支持向量机若干问题及应用研究[D].上海：上海大学,2008
[153]C.J.Xin, LibSVM:A library for Support Vector Machines[EB/OL]. http://www.csie.ntu.edu.tw/-cjlin/libsvm
[154]梁锦锦,刘三阳,吴德.空间支持向量域分类器[J].西安电子科技大学学报,2008,35(6)：1080-1089
[155]姜映映,田丰,王绪刚等.基于模板匹配和SVM的草图符号自适应识别方法[J],计算机学报,2009.2：175-182
[156]姜万录,吴胜强.基于SVM和证据理论的多数据融合故障诊断方法[J].仪器仪表学报.2010.8：1738-1743
[157]常甜甜.支持向量机学习算法若干问题的研究[D].西安：西安电子科技大学,2010
[158]王秀红,鞠时光.用于文本相似度计算的新核函数[J].通信学报,2012.12：43-48
[159]王秀红,鞠时光.基于混合核函数的分布式信息检索结果融合[J].通信学报, 2011.4：112-118
[160]Kernel functions for machine learning applications[EB/OL]. http://crsouza.blogspot.com/2010/03/kernel-functions-for-machine-learning
[161]TIAN Y, LI H, CAI Q, et al. Measuring the similarity of short texts by word similarity and tree kernels[A]. Proceedings of Conf YC-ICT'10[C]. New York, USA, 2010:1006-72
[162]邬啸,魏延,吴瑕.基于混合核函数的支持向量机[J].重庆理工大学学报(自然科学版),2011(10)：66-69
[163]YZhang, Y.C.Liu. Traffic forecasting using least squares support vector machines[J]. Transport metrical,2009,5(3):193-213
[164]杨全,彭进业.采用深度图像信息和SLVW的手语识别[J/OL],http://www.cnki.net/kcms/detail/11.2127.TP.20130618.1559.007.html,2013.6.18
[2]马继勇.手语理解的统计模型研究[D].北京：中国科学院计算技术研究所,2001
[3]郑重雨.基于运动特征的手语动画关键帧调整和插值方法的研究[D].北京：北京工业大学,2009
[4]中国聋人协会.中国手语(修订版)[M].北京：北京华夏出版社,2003.4
[5]李连昌.复杂场景下的视觉目标跟踪方法研究[D].济南：山东大学,2010
[6]段宏湘.序列图像中动态手势跟踪与建模方法研究[D].兰州：兰州理工大学,2011
[7]张毅,张姣,罗元.基于手势跟踪的智能轮椅控制系统[J].重庆邮电大学学报,2011.6：73-76
[8]翁汉良,战荫伟.基于视觉的多特征手势识别[J].计算机工程与科学,2012.2：123-127
[9]王超,崔更申,黄延辉等.线上线下结合的实时手势跟踪系统[J].桂林电子科技大学学报,2012.2：95-99
[10]王西颖,张习文,戴国忠.一种面向实时交互的变形手势跟踪方法[J].软件学报,2007.10：2423-2433
[11]Wang R, Popovic J. Real-Time Hand-Tracking with a Color Glove[J]. ACM Trans on Graphics,2009,28(3):1-8
[12]冯志全,杨波,李毅等.以时间优化为目标的粒子滤波手势跟踪方法研究[J].电子学报,2009.9：1989-1995
[13]常亚南.基于HMM的动态手势识别[D].华南理工大学,2012
[14]王西颖,戴国忠,张习文等.基于HMM-FNN模型的复杂动态手势识别[J].软件学报,2008.9：2302-2312
[15]彭娟春,顾立忠,苏剑波.基于Camshift和Kalman滤波的仿人机器人手势 跟踪[J].上海交通大学学报,2006.7：1161-1165
[16]张秋余,胡建强,张墨逸.基于区域生长的Mean shift动态变形手势跟踪算法[J].模式识别与人工智能,2010.4：580-585
[17]甘明刚,陈杰,王亚楠等.基于Mean Shift算法和NMI特征的目标跟踪算法研究[J].自动化学报,2010.9：1332-1336
[18]王鑫,唐振民.一种改进的基于Camshift的粒子滤波实时目标跟踪算法[J].中国图象图形学报,2010.10：1507-1514
[19]Zhou Ren, Junsong Yuan, Zhengyou Zhang. Robust Hand Gesture Recognition Based on Finger-Earth Mover's Distance with a Commodity Depth Camera [A]. MM'11[C], Scottsdale, Arizona, USA, November 28-December 1,2011:1093-1096
[20]雷海龙,杨胜,刘红新.融合图像轮廓矩和傅立叶描述子的手势识别[J].微电子学与计算机,2011.4：550-558
[21]张海燕,王博亮.基于改进的Snake模型的眼细胞彩色图像边缘检测算法研究[J].厦门大学学报(自然科学版),2005,03：76-80
[22]曹听燕,赵继印,李敏.基于肤色和运动检测技术的单目视觉手势分割[J].湖南大学学报(自然科学版),2011.1：78-83
[23]杨波,宋晓娜,冯志全等.复杂背景下基于空间分布特征的手势识别算法[J].计算机辅助设计与图形学学报,2010.10：1841-1848
[24]Abbas Cheddad, Joan Condel. A skin tone detection algorithm for an adaptive approach to steganography[J]. Signal Processing,2009,89(12):2456-2478
[25]高建坡,王煜坚,杨浩等.一种基于KL变换的椭圆模型肤色检测方法[J].电子与信息学报,2007(07)：1739-1743
[26]王鼎,沈辉,娄海涛.一种基于H-CrCb颜色空间的肤色检测算法研究[J].计算机科学,2012(39)：183-187
[27]张登康,郭太良,姚剑敏等.改进肤色识别参量和轮廓特征相结合的指尖检测[J].光电子技术,2012.3：185-189
[28]Argyros AA, Lourakis MIA. Real time tracking of multiple skin colored objects with a possibly moving camera [A]. In Proc. ECCV[C],2004.3:368-379
[29]包加桐,宋爱国,郭晏.基于SURF特征跟踪的动态手势识别算法[J].机器人, 2011.4：482489
[30]陈意,杨平,陈旭光.一种基于加速度特征提取的手势识别方法[J].传感技术学报,2012.8：1073-1078
[31]梁振.基于辐射投影变换和极半径不变矩的静态手势识别研究[D].山东科技大学,2009
[32]张墨逸.基于表观的二维手势识别方法研究[D].兰州：兰州理工大学,2010
[33]文学志,方巍,郑钰辉.一种基于类Haar特征和改进AdaBoost分类器的车辆识别算法[J].电子学报,Vol.39, No.5,2011.5:1121-1126
[34]王凯,于鸿洋,张萍.基于AdaBoost算法和光流匹配的实时手势识别[J].微电子学与计算机, Vol.29, No.4,2012.4:138-141
[35]Paul Doliotis, Alexandra Stefan, Christopher McMurrough, et al. Comparing Gesture Recognition Accuracy Using Color and Depth Information [A]. PETRA 2011[C], Crete, Greece,2011:1-7
[36]覃文军,吴成东,赵姝颖等.基于傅立叶描述子-BP神经网络的手势识别方法[J].东北大学学报(自然科学版),Vol.30, No.9,2009.9:1232-1235
[37]谭昶,肖南峰.基于改进RCE和RBF神经网络的静态手势识别[J].计算机工程与应用,2011.47(7)：172-176
[38]李文生,解梅,邓春健.一种快速的动态手势学习和识别方法[J].南京大学学报(自然科学版),2012.4：421-428
[39]Yun Liu, Yanmin Yin, Shujun Zhang. Hand Gesture Recognition Based on HU Moments in Interaction of Virtual Reality [A].4th International Conference on Intelligent Human-Machine Systems and Cybernetics[C],2012:145-148
[40]阳平,陈香,李云等.一种基于融合多传感器信息的手语手势识别方法[J].航天医学与医学工程,2012.4：276-281
[41]王德鑫,石崇林,张茂军.基于Petri网和BPNN的多重触控手势识别[J].模式识别与人工智能,2010.3：408-413
[42]孙荣斌.基于粒子滤波的手势识别技术研究[D].南京：南京理工大学,2009
[43]王文会,陈香,阳平等.基于多传感器信息检测和融合的中国手语识别研究[J].中国生物医学工程学报,2010.5：665-671
[44]ULRYCHJ, KOPECKY M. Visual similarity in sign language [A]. Proc. of the 24th International Conference on Data Engineering[C],2008:346-353
[45]Priyal S P, Bora P K. A study on static hand gesture recognition using moments[A]. Proc of IEEE SPCOM'10[C], Bangalore IEEE,2010:1-5
[46]P.Dreuw, J.Forster, H.Ney. Tracking Benchmark Databases for Video-Based Sign Language Recognition[A]. In ECCV International Workshop on Sign, Gesture and Activity(SGA)[C]. Crete, Greece, September 2010:335-341
[47]P.Dreuw, D.Rybach, T.Deselaers, et al. Speech Recognition Techniques for a Sign Language Recognition System[R]. In Interspeech, Antwerp, Belgium, August 2007. ISCA best student paper award Interspeech,2007
[48]吴江琴,高文,陈熙霖等.基于ANN/HMM的中国手语识别系统[J].计算机工程与应用,1999.9：1-4
[49]Wen Gao, Ji-Yong Ma, Jiang-Qin Wu, et al. Large vocabulary sign language recognition based on HMM/ANN/DP[J]. International Journal of Pattern Recogniton and Artificial Intelligence,2000,14(5):587-602
[50]杨筱林,姚鸿勋.基于多尺度形状描述子的手势识别[J].计算机工程与应用,2004,40(32)：76-78
[51]张良国,高文,陈熙霖等.面向中等词汇量的中国手语视觉识别系统[J].计算机研究与发展,2006,43(3)：476-482
[52]姜峰,高文,姚鸿勋等.非特定人手语识别问题中的合成数据驱动方法[J].计算机研究与发展,2007,44(5)：873-881
[53]Zhu Yuanxin, XuGuangyou, Huang Yu. Appearance-based dynamic hand gesture recognition from image sequences with complex background[J]. Journal of Software, 2000,11(1):54-61
[54]ZHANG Shi-lin, ZHANG Bo. Using HMM to sign language video retrieval[A]. Proc of the 2nd International Conference on Computational Intelligence and Natural Computing[C].2010:55-59
[55]Fels. Glove talk II:Mapping hand gestures to speech using neural networks-an approach to building adaptive interfaces[D]. Computer Science Department, University of Torino,1994
[56]T.Takahashi, F.K.Shino. Hand gesture coding based on experiments using a handgesture interface device[J]. SIGCHI Bulletin,1991,23(2):67-73
[57]Liang R H, Ouhyoung M, A real-time continuous gesture recognition system for sign language[A]. Proc of 3rd International Conference on Automatic Face and Gesture Recognition[C]. Nara, Japan,1998:558-565
[58]Y Cui, D L Swets, J JWeng. Learning-based hand sign language using SHOSLIF-M[A]. Proc of ICCV'96[C].1996:621-636
[59]Triesch J, MalsburgC.Robust Classification of Hand Postures Against Complex Background[A]. International Conference on Automatic Face and Gesture Recognition[C].1996:170-175
[60]Quek F, Zhao M. Inductive Learning in Hand Pose Recognition[J], IEEE Automatic Face and Gesture Recognition,1996:78-83
[61]Davis, ShahParallel. Hidden Markov Models for American Sign Language Recognition[A]. Proceedings of the International Conference on Computer Vision[C]. Kerkyra, Greece, September,1999:22-25
[62]Starner T, Pentland A. Real-time American sign language recognition from video using hidden Markov models[R]. Media Lab, MIT:Technical Report TR375,1996
[63]Christian Vogler, Dimitris Metaxas. Adapting hidden Markov models for ASL recognition by Using three-dimensional computer vision methods[A]. Proceedings of the IEEE International Conference On Systems, Man and Cybernetics[C].Orlando, FL, 1997:156-161
[64]ChristianVogler, Dimitris Metaxas. ASL recognition based on a coupling between HMMS and 3D motion analysis[A]. Proceedings of the IEEE International Conference on Computer Vision[C].Mumbai, India,1998:365-371
[65]ChristianVogler, Dimitris Metaxas. Toward scalability in ASL Recognition: Breaking Down Signs into Phonemes[A]. Proceedings of Gesture Workshop[C]. Gif-sur-Yvette, France,1999:400-404
[66]ChriatianVogler, DimitrisMetaxa. Parallel hidden Markov models for American Sign Language recognition[A]. Proceedings of the International Conference on Computer Vision[C].Kerkyra, Greece,1999:224-228
[67]J.W. Deng, H. T. Tsui. A two-step approach based on HMM for the recognition of ASL[A]. The 5th Asian Conf. Computer Vision[C]. Melbourne, Australia, 2002:363-369
[68]CHEN Qing, GEORGANAS N D, PETRIU E M. Real-time vision-based hand gesture recognition using Haar-like features[A]. Proc. of IEEE Instrumentation and Measurement Technology Conference[C]. IEEE Press,2007:1-6
[69]SILANON K, SUVONVORN N. Hand motion analysis for Thai alphabet recognition using HMM[J]. International Journal of Informationand Electronics Engineering,2011,1(1):65-71
[70]Ong.Eng-Jon, Cooper.Helen, Pugeault.Nicolas, et al. Sign Language Recognition using Sequential Pattern Trees[A]. The IEEE Conference on Computer Vision and Pattern Recognition(CVPR) [C]. Providence, Rhode Island,16-21 June, 2012:2200-2207
[71]NIEBLES J, WANG H, LI F F. Unsupervised learning of human action categories using spatial temporal words [J]. International Journal of Computer Vision, 2008,79(3):299-318
[72]DARDAS N H, PETRIU E M. Hand gesture detection and recognition using principal component analysis[A]. Proceedings of International Conference on Computational Intelligence for Measurement Systems and Applications (CMS A) [C]. Ottawa:IEEE,2011:1-6
[73]黄康泉,陈壁金,郑博等.Kinect在视频会议系统中的应用[J].广西大学学报：自然科学版,2011,36：308-310
[74]Zheng Xiao, Fu Mengyin, Yang Yi et al.3D Human Postures Recognition Using Kinect[A].4th International Conference on Intelligent Human-Machine Systems and Cybernetics (IHMSC2012) [C]. Jiang Xi, Nan Chang, P.R.China, 2012.8.26-27:1796-1801
[75]RAHEJA J L, CHAUDHARV A, SIGNAL K. Tracking of fingertips and centers of palm using Kinect[A]. Proc. of the 3rd International Conference on Computational Intelligence, Modeling and Simulation[C]. IEEE Press,2011:248-252
[76]齐苏敏.序列图像中手势跟踪与识别技术的研究[D].苏州：苏州大学,2008
[77]Lipton A.J, Fujiyoshi H., Patti R.S..Moving target classification and tracking from real-time video[A]. Proceedings of IEEE Workshop on Applications of ComputerVision[C].Princeton, NJ, USA,1998:3642-3647
[78]刘亚辉.面向智能空间的多视角视觉系统关键技术研究[D].北京：北京邮电大学,2011
[79]Jin Li, JunhongXu,Wang Cong, et al. Research on real-time detection of moving target using gradient optical flow[A]. Proceedings of IEEE International Conference on Mechatronics and Automation[C].Niagara Fails, CanMa,2005,4:1796-1801
[80]McKenna S., J. JabriS, Duric Z, et al. Tracking groups of people[J]. Computer Vision and Image Understanding.2000,80(1):42-56
[81]Yamazawa K, Yokoya N. Detecting moving objects from omnidirectional dynamic images based on adaptive background subtraction[A]. Proceedings of International Conference on Image Processing[C].2003,3:953-956
[82]Stauffer C, Grimson W E L. Adaptive background mixture models for realtime tracking[A]. Computer Vision and Pattern Recognition[C], IEEE Computer Society Conference,1999,2:8-14
[83]R.Collin. A system for video surveillance and monitoring VSAM final report[R].Carnegie Mellon University,Technical Report:CMU-IR-TR-00-12,2000
[84]Paul Doliotis, Alexandra Stefan, Christopher McMurrough, et al. Comparing Gesture Recognition Accuracy Using Color and Depth Information[A]. PETRA 2011 [C]. Crete, Greece:20-23
[85]P. Perez, J. Vemaak and ABlake.Data fusion for visual tracking with particle[A]. Proceedings of the IEEE[C].2004,92(3):495-513
[86]R.T.Collies, Y.X.Liu and M.Leordeanu. Online selection of discriminative tracking features[A].IEEE Transactions on Pattern Analysis and Machine Intelligence[C].2005,27(10):1631-1643
[87]E.Wan and R.Van der Mexwe. The unscented Kalman Filter[A].S.Hayhin(Eels.), Kalman Filtering and Neural Networks[C].John Wiley & Sons Inc,2001:221-280
[88]K.Nummiaro, E.Koller-Meier, L.Van GOOL. An adaptive color-based particle filter[J]. Image and Vision Computing.2003,21(1):99-110
[89]任仙怡,廖云涛,张桂林等.一种新的相关跟踪方法研究[J].中国图形图像学报.2002.7(6)：553-557
[90]Hager GD, Dewan M, Stewart C V. Multiple Kernel Tracking with SSD[A]. Proceedings of IEEE Conference on Computer Vision and Pattern Recognition[C].2004:790-797
[91]C.Yang, R.Duraiswami, L.Davis. Efficient mean-shift tracking via a new similarity measure[A]. Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition[C].San Diego, USA,2005:176-183
[92]Zhaowen Wang, Xiaokang Yang, Yi Xu, et al. CamShift guided particle filter for visual tracking[J]. Pattern Recognition Letters.30(2009):407-413
[93]J.Kovacevic, S.Juric-Kavelj, I.Petrovic, et al. An Improved CamShift Algorithm Using Stereo Vision For Object Tracking[A]. MIPRO 2011[C]. Opatija, Croatia, May 23-27,2011:707-710
[94]Kinect[EB/OL]. http://en.wikipedia.org/wiki/Kinect,2013.3
[95]Windows Kinect SKD[EB/OL]. http://kinectforwindows.org,2013.3
[96]Prime Sense OpenNI[EB/OL]. http://www.primesense.com,2013.3
[97]罗元,谢或,张毅.基于Kinect传感器的智能轮椅手势控制系统的设计与实现[J].机器人ROBOT. Vol.34, No.l, Jan.,2012,34(1):110-113
[98]文志强.基于均值偏移算法的运动目标跟踪技术的研究[D].长沙：中南大学,2008
[99]王帅.一种基于MeanShift算法的目标跟踪算法的研究[D].济南：山东大学,2011
[100]Otsu N. A Threshold Selection Method from Gray-level Histograms[J]. IEEE Trans on SMC.2009(1):62-69
[101]CARREIRA PERPINAN M A. Acceleration strategies for Gaussian Mean Shift image segmentation[A]. Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition[C]. New York,2006:543-549
[102]赵雪.基于自适应颜色特征的CamShift改进算法的研究[D].大连：大连海事大学,2011
[103]胡铟,杨静宇.基于分块颜色直方图MeanShift跟踪算法[J].系统仿真学报.2009,10：2936-2939
[104]何革.基于MeanShift算法的视频目标追踪研究[D].重庆：重庆大学,2010
[105]于永健,王向阳,吴俊峰.基于颜色复杂度的加权颜色直方图图像检索算法[J].小型微型计算机系统.2009,Vol-30.No.3:507-511
[106]Geetha M, Manjusha U C. A Vision Based Recognition of Indian Sign Language Alphabets and Numerals Using B-Spline Approximation[J]. International Journal on Computer Science and Engineering, Vol.4 No.03 March 2012:578-582
[107]RAHEJA J.L, CHAUDHARV A, SIGNAL K. Tracking of Fingertips and Centres of Palm using Kinect[A]. CIMSIM[C], Langkawi, IEEE,2011:248-252
[108]OpenKinect Organization. Imaging Information for Kinect[EB/OL]. http://openkinect.org/wiki/Imaging_Information,2013
[109]陈锻生,刘政凯.肤色检测技术综述[J].计算机学报,2006,29(2)：194-207
[110]曹听燕,赵继印,李敏.基于肤色和运动检测技术的单目视觉手势分割[J].湖南大学学报(自然科学版).Vlo.38, No.1,2011.1:78-83
[111]吴海滨,周后伟,张铁译等.基于OTSU的动态结合全局阈值的图像分割[J].大气与环境光学学报.Vol.7, No.6,2012.11:463-465
[112]范九伦,赵凤.灰度图像的二维Otsu曲线阈值分割法[J].电子学报.2007,4：751-755
[113]刘春生,常发亮,陈振学等.改进的高斯肤色模型及其在人脸检测中的应用[J].仪器仪表学报.2012,33(5)：1117-1121
[114]朱志亮,刘富国,陶向阳.基于积分图和粒子群优化的肤色分割[J].计算机工程与应用.2013.1：789-792
[115]郎咸朋,朱枫,都颖明等.基于积分图像的快速二维Otsu算法[J].仪器仪表学报,2009,30(1)：39-43
[116]Jang H, Do J H, Jung J, et al. View-invariant hand-posture recognition for soft-remo control system[A]. IEEE International Conference on Intelligent Robots and Systems[C], Sendai, Japan:IEEE,2004,1:295-300
[117]J.Zaletelj, J.Perhavc, J.F. Tasic. Vision-based Human-computer Interface Using Hand Gestures[A].2007 International Workshop on Image Analysis for Multimedia Interactive Services[C].Santorini,Greece,2007:1-4
[118]J.F.Xu, S.Q.Jun, Y.W.Shi, et al. Implementation of Digital Chime-bell Interaction System Driven by Hand Gesture[A]. The 2008 Interactional Conference on Pervasive Computing and Applications[C].Alexandria, Egypt,2008:264-268
[119]朱建伟.基于多摄像头的手势识别研究与实现[D].广州：中山大学,2010.6
[120]王骐,陈熙霖,王春立等.一种可处理数据缺失的视角无关手语识别方法[J].计算机学报,2009.5：954-960
[121]华斌,夏利娜.基于中值滤波和Hu矩向量的手语识别[J].计算机工程与设计.2011.2：615-618
[122]倪训博,赵德斌,高文.非特定人手语数据生成及其有效性检测[J].软件学报,2010.5：151-163
[123]汪道寅.基于SIFT图像配准算法的研究[D].合肥：中国科学技术大学,2011.5
[124]Morel J M, Yu G S. ASIFT:A New Framework for Fully Affine Invariant Image Comparison[J]. Society for Industrial and Applied Mathematics Journal on Image Sciences,2009,2(2):438-469
[125]Song R, Szymanski J. Well-distributed SIFT features[A]. Electronics Letters[C], 2009,45(6):308-310
[126]Zhou H Y, Yuan Y, Shi C M. Object tracking using SIFT features and mean shift[J]. Computer Vision and Image Understanding,2009,113(3):345-352
[127]蔺海峰,马宇峰,宋涛.基于SIFT特征目标跟踪算法研究[J].自动化学报,2010.8：1204-1208
[128]Juan L,Gwun O. A Comparison of SIFT, PCA-SIFT and SURF[J]. International Journal of Image Processing,2009,3(4):143-152
[129]陈小丹,杜宇人,高秀斌.一种基于SURF的图像特征点快速匹配算法[J]. 扬州大学学报(自然科学版),2012.4：15-18
[130]Y Bastanlar,A Temizel,Y Yardimci.Improved SIFT matching for image pairs with scale difference[J]. Electronics Letters,2010.6:346-348
[131]Luo Juan.A comparison of sift, pea-sift and surf[J]. International Journal of Image Processing,2009.12:143-152
[132]范新南,丁朋华,刘俊定等.融合灰度和SURF特征的红外目标跟踪[J].中国图象图形学报,2012.11：1376-1383
[133]卢振坤,玉振明.基于自适应局部Log-Gabor能量的多聚焦图像融合[J].电子学报,2013.1：35-39
[134]LI X, He M, Roux M. Multifocus image fusion based on redundant wavelet transform[J]. IET Image Processing,2010:283-293
[135]吴霁轩.基于二维Gabor特征和多特征融合的人脸识别研究[D].长春：吉林大学,2012
[136]耿佳佳,康晓东,王昊等.联合Gabor和分形的医学影像图像局部特征描述[J].计算机科学,2012,S3：97-101
[137]周健,王华彬,陶亮.基于离散傅里叶变换和块时间递归并行格型结构的离散Gabor分析窗求解[J].电子学报,2012.9：1839-1843
[138]王向阳,杨红颖,郑宏亮等.基于视觉权值的分块颜色直方图图像检索算法[J].自动化学报,2010.10：1489-1492
[139]陈海林,吴秀清.基于双空间金字塔匹配核的图像目标分类[J].中国科学技术大学学报,2010(03)：313-320
[140]贾世杰,孔祥维.一种新的直方图核函数及在图像分类中的应用[J].电子与信息学报,2011.7：1738-1742
[141]王宇石；高文.用基于视觉单词上下文的核函数对图像分类[J].中国图象图形学报,2010.4：607-616
[142]刘扬闻,霍宏,方涛.词包模型中视觉单词歧义性分析[J].计算机工程,Vol.37, No.19,2011.10:164-170
[143]张秋余,王道东,张墨逸等.基于特征包支持向量机的手势识别[J].计算机应用,2012,32(12)：3392-3396
[144]陆福相.基于内容的场景和物体类别识别研究[D].上海：上海交通大学,2011
[145]李新德,金晓彬,张秀龙.一种基于BoW物体识别模型的视觉导航方法[J].东南大学学报(自然科学版).2012.3：3-8
[146]武京相.融合全局和局部特征的医学图像分类[D].成都：电子科技大学,2010
[147]王科俊,邹国锋,阎涛.融合全局与局部特征的单样本人脸识别[J].中南大学学报(自然科学版),2011.9：1-7
[148]庄凌,庄越挺,吴江琴等.一种基于稀疏典型性相关分析的图像检索方法[J].软件学报,2012.5：89-94
[149]HAI Xianwang.Local two-dimensional canonical correlation analysis[J]. Signal Processing.2010.7:921-924
[150]刘瑞瑞.人脸特征提取及融合方法研究[D].南京：东南大学,2010
[151]梁锦锦.支持向量机和支持向量域描述的若干问题研究[D].西安：西安电子科技大学,2009
[152]彭新俊.支持向量机若干问题及应用研究[D].上海：上海大学,2008
[153]C.J.Xin, LibSVM:A library for Support Vector Machines[EB/OL]. http://www.csie.ntu.edu.tw/-cjlin/libsvm
[154]梁锦锦,刘三阳,吴德.空间支持向量域分类器[J].西安电子科技大学学报,2008,35(6)：1080-1089
[155]姜映映,田丰,王绪刚等.基于模板匹配和SVM的草图符号自适应识别方法[J],计算机学报,2009.2：175-182
[156]姜万录,吴胜强.基于SVM和证据理论的多数据融合故障诊断方法[J].仪器仪表学报.2010.8：1738-1743
[157]常甜甜.支持向量机学习算法若干问题的研究[D].西安：西安电子科技大学,2010
[158]王秀红,鞠时光.用于文本相似度计算的新核函数[J].通信学报,2012.12：43-48
[159]王秀红,鞠时光.基于混合核函数的分布式信息检索结果融合[J].通信学报, 2011.4：112-118
[160]Kernel functions for machine learning applications[EB/OL]. http://crsouza.blogspot.com/2010/03/kernel-functions-for-machine-learning
[161]TIAN Y, LI H, CAI Q, et al. Measuring the similarity of short texts by word similarity and tree kernels[A]. Proceedings of Conf YC-ICT'10[C]. New York, USA, 2010:1006-72
[162]邬啸,魏延,吴瑕.基于混合核函数的支持向量机[J].重庆理工大学学报(自然科学版),2011(10)：66-69
[163]YZhang, Y.C.Liu. Traffic forecasting using least squares support vector machines[J]. Transport metrical,2009,5(3):193-213
[164]杨全,彭进业.采用深度图像信息和SLVW的手语识别[J/OL],http://www.cnki.net/kcms/detail/11.2127.TP.20130618.1559.007.html,2013.6.18