[1]王原,汤勇明,王保平.基于混合高斯模型的非固定握持姿势手势识别[J].东南大学学报(自然科学版),2014,44(2):239-243.[doi:10.3969/j.issn.1001-0505.2014.02.003]
 Wang Yuan,Tang Yongming,Wang Baoping.Gesture recognition with unfixed holding position based on Gaussian mixture model[J].Journal of Southeast University (Natural Science Edition),2014,44(2):239-243.[doi:10.3969/j.issn.1001-0505.2014.02.003]
点击复制

基于混合高斯模型的非固定握持姿势手势识别()
分享到:

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
44
期数:
2014年第2期
页码:
239-243
栏目:
电气工程
出版日期:
2014-03-20

文章信息/Info

Title:
Gesture recognition with unfixed holding position based on Gaussian mixture model
作者:
王原汤勇明王保平
东南大学电子科学与工程学院, 南京210096
Author(s):
Wang Yuan Tang Yongming Wang Baoping
School of Electronic Science and Engineering, Southeast University, Nanjing 210096, China
关键词:
手势识别 混合高斯模型 用户体验
Keywords:
gesture recognition Gaussian mixture model user experience
分类号:
TM391.4
DOI:
10.3969/j.issn.1001-0505.2014.02.003
摘要:
针对手势识别研究中普遍要求用户以严格固定方式握持数据采集设备,致使用户体验差的问题,使用混合高斯模型(Gaussian mixture model,GMM)对非固定握持姿势的手势识别算法进行改进,以提高手势人机交互时的舒适性.首先通过GMM从加速度传感器数据中提取用户握持姿势数据,然后借助握持信号实现手势命令数据提取与坐标转换,使识别系统能够自适应不同的握持姿势.为使GMM可以同时满足手势识别应用中对稳定性和适应速度的要求,优化了GMM的学习机制,包括增加备则模态和改善优先级计算.实验结果表明,所述系统在滚转角和俯仰角+60°~-60°、偏摆角+20°~-20°范围内,握持姿势对手势识别正确率没有明显影响,实现了非固定握持姿势的手势识别,起到了提高用户体验的作用.
Abstract:
In most gesture recognition researches, participants are asked to hold the data collecting device in fixed position strictly, which causes poor user experience. To solve this problem, an enhancement algorithm based on the Gaussian mixture model(GMM)is presented. It can achieve gesture recognition with unfixed holding position and improve the interaction comfortability. First, the holding position information is abstracted from raw acceleration data by the GMM. Then the coordinate transformation is conducted and the gesture operating information is separated with the holding position information. To meet the requirements for stability and recognition speed, the parameter updating strategy of the GMM is improved by adding backup component and optimizing the priority consideration. The experimental results show that when the roll angle and the pitching angle are between -60° to +60°, the yaw angle is between -20° to +20°, the holding position has no significant impact on recognition accuracy. So the gesture recognition algorithm is improved without fixed holding position, thus achieve better user experience.

参考文献/References:

[1] Akl A, Feng C, Valaee S. A novel accelerometer-based gesture recognition system[J]. IEEE Transactions Signal Processing, 2011, 59(12): 6197-6205.
[2] Popa M. Hand gesture recognition based on accelerometer sensors[C]//Proceedings of 7th Intl Conf on Networked Computing and Advanced Information Management. Gyeongju, Republic of Korea, 2011: 115-120.
[3] Schlömer T, Poppinga B, Henze N, et al. Gesture recognition with a Wii controller[C]//2nd Intl Conf on Tangible and Embedded Interaction. New York, USA, 2008: 11-14.
[4] Pylvänäinen T. Accelerometer based gesture recognition using continuous HMMs[M]//Pattern Recog Image Anal. Berlin, German: Springer, 2005:639-646.
[5] Liu J, Zhong L, Wickramasuriya J, et al. Uwave: accelerometer-based personalized gesture recognition and its applications[J]. Pervasive Mobile Compute, 2009, 5(6): 657-675.
[6] Hartmann B, Link N. Gesture recognition with inertial sensors and optimized DTW prototypes[C]//IEEE International Conference on Systems Man and Cybernetics. Istanbul, Turkey, 2010:2102-2109.
[7] Hussain Shah M A. User independent hand gesture recognition by accelerated DTW[C]//International Conference on Informatics, Electronics and Vision. Dhaka, Bangladesh, 2012: 1033-1037.
[8] Xu R Z, Zhou S L, Li W J. MEMS accelerometer based nonspecific-user hand gesture recognition[J]. IEEE Sensors Journal, 2012, 12(5): 1166-1173.
[9] 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, 25(8): 1073-1078.
  Chen Yi, Yang Ping, Chen Xuguang. A gesture recognition method based on acceleration feature extraction[J]. Chinese Journal of Sensors and Actuators, 2012, 25(8): 1073-1078.(in Chinese)
[10] Stauffer C, Grimson W E L. Adaptive background mixture models for real-time tracking[C]//Computer Vision and Pattern Recognition. Fort Collins, USA, 1999, 2:246-252.
[11] 杜鉴豪. 监控视频中的人体异常行为检测研究[D]. 杭州:浙江大学电气工程学院,2010.
[12] Zhou S L, Shan Q, Fei F, et al. Gesture recognition for interactive controllers using MEMS motion sensors[C]//IEEE Intl Conf Nano/Micro Engineered and Molecular Systems.Shenzhen, China, 2009: 935-940.

相似文献/References:

[1]胡春华,马旭东,戴先中,等.一种基于标准混合高斯模型的快速人脸检测方法[J].东南大学学报(自然科学版),2007,37(3):389.[doi:10.3969/j.issn.1001-0505.2007.03.007]
 Hu Chunhua,Ma Xudong,Dai Xianzhong,et al.Method for fast face-detection based on normalization Gaussian mixture model[J].Journal of Southeast University (Natural Science Edition),2007,37(2):389.[doi:10.3969/j.issn.1001-0505.2007.03.007]

备注/Memo

备注/Memo:
收稿日期: 2013-09-27.
作者简介: 王原(1986—),男,硕士生;汤勇明(联系人),男,博士,研究员,tym@seu.edu.cn.
基金项目: 国家高技术研究发展计划(863计划)资助项目(2012AA03A302)、高等学校学科创新引智计划资助项目(B07027).
引用本文: 王原,汤勇明,王保平.基于混合高斯模型的非固定握持姿势手势识别[J].东南大学学报:自然科学版,2014,44(2):239-243. [doi:10.3969/j.issn.1001-0505.2014.02.003]
更新日期/Last Update: 2014-03-20