液晶与显示, 2014, 29 (1): 129, 网络出版: 2014-03-14  

改进型高精度实时视频多指尖检测方法的研究

Improved high-accuracy multiple fingertips detection method for real-time video
作者单位
1 福州大学 物理与信息工程学院,福建 福州 350002
2 TCL 集团工业研究院 新型显示技术与系统项目组,广东 深圳 518052
摘要
指尖提取已成为基于视觉的裸手人机交互的关键技术,针对目前很多识别手势和指尖检测方法存在的对设备要求高、系统效率低、局限于单指检测、应用局限性高和对于弯曲手指存在误检等诸多问题,提出了一种改进型高精度曲率多指尖检测法。首先,通过对提取到的手势边缘序列点求曲率和像素值判断法排除手指凹槽。同时还提出了掌心测距法结合手掌外接圆法等有效地去除了握拳时和手指半弯时解决类指尖误判问题的方案。最后,整合优化算法流程,实验验证表明该改进的检测方法能较好地提高程序运行效率,检测精度与运行速度得到了明显的改善。
Abstract
Fingertip extraction has become the key technology of bare hand human-computer interaction based on vision. Aiming at solving the high requirements for equipment,low system efficiency, limited to signle finger detection,high application limitations and error detection of bending fingers, and many other problems of many hand gesture recognition methods and fingertip detection methods at present,this article proposes a modified high precision multiple fingertips detection method based on curvature.Firstly,it eliminates the finger grooves by the curvature of extracted edge gesture sequence points and pixel values judgment method.And it proposes the palm ranging method and the inscribed circle of palm method to effectively wipe out the class fingertips error judgement of fist and bending fingers at the same time. Finally, it optimizates algorithm flow chart.Experiments show that the improved detection method has improved the efficiency of program and obviously improved the detection precision.
参考文献

[1] De Dios J J,Gareia N.Face detcection based on a new color space YCgCr [C].Proceedings of International Conference on Image Processing,2003:909-912.

[2] 蔡祖洪.复杂背景及复杂运动背景下的指尖检测方法[J].杭州电子科技大学学报.2012,32(1):60-63.

    Cai Z H.Fingertip detection approach from cluttered background and moving foreground [J].Journal of Hangzhou Dianzi University,2012,32(1):60-63.(in Chinese)

[3] 于蕴杰,管业鹏,熊会军.指势识别的实时指尖提取[J].计算机工程与应用,2011,47(6):219-222.

    Yu Y J, Guan Y P, Xiong H J. Real-time extraction of fingertip for pointing gesture recognition [J]. Computer Engineering and Applications,2011,47(6):219-222.(in Chinese)

[4] Cristina M,Javier V,Ramon M, et al.Real-time hand tracking and gesture recognition for human computer interaction [C].Electronic Letters on Computer Vision and Image Analysis,Spain:cvc press,2000: 1-7.

[5] Le D,Mizukawa M.Fast fingertips positioning based on distance-based feature pixels [C].Third International Conference of Communication and Electronics,UK:IEEE,2010: 184-189.

[6] Wen X T, Niu Y Y. A method for hand gesture recognition based on morphology and fingertip-angle [C].The 2nd International Conference of Computer and Automation Engineering(ICCAE),UK:IEEE,2010: 688-691.

[7] 苑玮琦,朱春艳,柯丽.基于手指相对长度的手形识别方法[J].光电子·激光,2009(5):685-689.

    Yuan W Q, Zhu C Y, Ke L. Hand shape identification method based on finger relative length [J].Journal of Optoelectronics·Laser,2009(5):685-689.(in Chinese)

[8] Kenii O, Yoichi S.Hideki K.Real-time tracking of multiple fingertips and gesture recognition for augmented desk interface systems [J].IEEE Computer Graphics and Applications,2002, 22(6): 64-71.

[9] 何东健.数字图像处理[M]. 2版. 西安:西安电子科技大学出版社,2008:186-188.

    He D J.Digital Image Processing[M].2nd ed.Xi’an: Xi’an University of Electronic Science and Technology Press, 2008:186-188.(in Chinese)

[10] Zhang X, Li D L, Yang W Z,et al. An improved morphological edge detection method for color images of cotton foreign fibers [J].Sensor Letters, 2011, 9(3): 1020-1023.

[11] Wang Y,Yuan J H,Chang S J,et al. Gesture labelingbased on gaze direction recognition for human-machine interaction [J].Optical Engineering,2002,41(8):1840-1844.

[12] 磨少清,刘正光,张军,等.基于图像自身的图像边缘检测阈值自动设定方法[J].光电子· 激光,2011, 22(8): 1246-1250.

    MO S Q,Liu Z G, Zhang J,et al. An unsupervised threshold selection method for edge detection based on image self-information [J].Journal of Optoelectronics·Laser,2011, 22(8): 1246-1250.(in Chinese)

[13] 李博男, 林凡. 基于曲率的指尖检测方法[J].南京航空航天大学学报,2012,44(4):587-591.

    Li B N,Lin F. Fingertip detection method based on curvature[J].Journal of Nanjing University of Aeronautics&Astronautics,2012,44(4):587-591.(in Chinese)

[14] Oka K, Sato Y, Koike H. Real-time tracking of multiple fingertips and gesture recognition for augumented desk interface system [C].Proceeding of the 5th IEEE Intl Conf on Automatic Face and Gesture Recognition,UK:IEEE,2002:411-416.

[15] 杨端端,金连文,尹俊勋.手指书写汉字识别系统中的指尖检测方法[J].华南理工大学学报,2007, 35(1): 58-63.

    Yang D D,Jin L W,Yin J X.Fingertip detection approach for finger writing Chinese character recognition system[J].Journal of South China University of Technology,2007,35(1):58-63.(in Chinese)

[16] Weissmann J,Salomon R.Gesture recognition for virtual reality applications using data gloves and neural networks[J].International Joint Conference on Neural Networks(IJCNN’99),1999,(3):2043-2046.

[17] Hyung-O K, Kim S,Sung-Kee.Pointing gesture-based unknown object extraction for learning objects with robot[C]. Proceedings of International Conference on Control Automation and Systems,UK:IEEE,2008: 2156-2161.

黄阿云, 郭太良, 林志贤, 姚剑敏. 改进型高精度实时视频多指尖检测方法的研究[J]. 液晶与显示, 2014, 29(1): 129. HUANG A-yun, GUO Tai-liang, LIN Zhi-xian, YAO Jian-min. Improved high-accuracy multiple fingertips detection method for real-time video[J]. Chinese Journal of Liquid Crystals and Displays, 2014, 29(1): 129.

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!