基于预分类处理的实时多手势识别算法

黎明聪; 吴逸畅; 张锡斌; 曾志远; 王嘉辉

doi:10.3788/YJYXS20173204.0308

您当前的位置：

首页 >

文章列表页 >

基于预分类处理的实时多手势识别算法

显示与图像 | 更新时间：2020-07-07

- 基于预分类处理的实时多手势识别算法
- Method of real-time multiple gestures recognition based on pre-classification
- 液晶与显示 2017年32卷第4期页码：308-315
- 作者机构：
  
  1.中山大学物理学院, 广东广州 510275
  2.中山大学光电材料与技术国家重点实验室, 广东广州 510275
- 作者简介：
  
  [ "黎明聪 (1995-), 男, 广东江门人, 主要从事图像处理, 人机交互方面的研究。E-mail:limc3@mail2.sysu.edu.cn" ]
  王嘉辉 (1979-), 男, 广东广州人, 博士, 实验师, 2009年于中山大学获得硕士学位, 2014年于中山大学获得博士学位, 主要从事3D显示、图像处理方面研究。E-mail:wangjh@mail.sysu.edu.cn
- 基金信息：
- DOI：10.3788/YJYXS20173204.0308
  中图分类号： TP394.1;TH691.9
- 收稿日期：2016-10-20，
  
  录用日期：2016-12-12，
  
  纸质出版日期：2017-04
- 稿件说明：
移动端阅览
黎明聪, 吴逸畅, 张锡斌, 等. 基于预分类处理的实时多手势识别算法[J]. 液晶与显示, 2017,32(4):308-315.

Ming-cong LI, Yi-chang WU, Xi-bin ZHANG, et al. Method of real-time multiple gestures recognition based on pre-classification[J]. Chinese journal of liquid crystals and displays, 2017, 32(4): 308-315.
黎明聪, 吴逸畅, 张锡斌, 等. 基于预分类处理的实时多手势识别算法[J]. 液晶与显示, 2017,32(4):308-315. DOI： 10.3788/YJYXS20173204.0308.

Ming-cong LI, Yi-chang WU, Xi-bin ZHANG, et al. Method of real-time multiple gestures recognition based on pre-classification[J]. Chinese journal of liquid crystals and displays, 2017, 32(4): 308-315. DOI： 10.3788/YJYXS20173204.0308.

摘要

为了改善常见的背景模型在手势分割易受环境因素影响的缺点，以及提高基于傅里叶描述子的神经网络对大量手势分类的识别率，本文提出一种采用预分类的综合手势识别算法。首先，分割部分使用背景差分检测出手部，同时利用手分类器对手部识别手掌位置，获取感兴趣区域，同时对手和背景分别进行掩膜更新，进入下一循环，具有较高的分割效果。接着，利用指尖检测和傅里叶描述子对区域进行特征提取。识别过程先根据指尖数目及实现功能进行预分类，再利用神经网络对傅里叶特征系数进行实时的识别并返回识别结果。最后，进行与直接的傅里叶系数的识别比较及复杂背景下与传统背景差分方法识别率比较，结果表明该算法普通手势比直接分类平均识别率高4.162%，易误检手势识别率提高最高达91.7%；复杂背景下性能也有较大改善。鲁棒性较强，可以容纳大量手势定义，满足人机交互的要求。

Abstract

To improve the common background model that can be easily faulted by environmental factors

and the efficiency of neural network based on Fourier descriptor for a large number of gesture recognition

a pre-classified gesture recognition algorithm is designed. First

background difference detection plus a hand classifier is employed to recognize palm position and obtain regions of interest

while masks are updated respectively in hand regions and background. Experiments indicate its remarkable segmentation effects. Then

fingertips detection follows and Fourier descriptors are extracted from areas of interest. Identification process includes the preliminary classification dependent on the number of fingertips detected as well as the potential movement. Then a neural network based on Fourier coefficients is to process the extracted feature data and feedback the results. In the end

comparison shows that this algorithm involving pre-classification system performs better in recognition. Ordinary gestures improve 4.162% in average

while easy-mistaken gestures escalate by up to 91.7%. Under complex environment

improvement can be seen compared to a common background difference method. Strong robustness and multiple gestures definition satisfy the requirement of real-time human-computer interaction.

关键词

Keywords

references

杨晓东, 陈益强, 于汉超, 等.面向可穿戴设备的超声波手势识别方法[J].计算机科学, 2015, 42(10):20-24.

YANG X D, Chen Y Q, YU H C, et al. Ultrasonic waves based gesture recognition method for wearable equipment[J]. Computer Science, 2015, 42(10):20-24. (in Chinese)

徐力群, 吴晓娟.基于颜色概率模型的实时手势跟踪算法[J].计算机工程与科学, 2005, 27(10):44-47.

XU L Q, WU X J. A real-time hand tracking algorithm based on the colour probability model[J]. Computer Engineering & Science, 2005, 27(10):44-47. (in Chinese)

STAUFFER C, GRIMSON W E L. Adaptive background mixture models for real-time tracking[C]//Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Fort Collins, CO:IEEE, 1999, 2:246-252.

STAUFFER C, GRIMSON W E L. Learning patterns of activity using real-time tracking[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2000, 22(8):747-757.

江超, 艾矫燕.基于OpenCV的摄像头动态手势轨迹识别及其应用[J].计算机应用, 2012, 32(S1):128-133.

JIANG C, AI J Y. Dynamic gesture trajectory recognition and its application design for camera based on OpenCV[J]. Journal of Computer Applications, 2012, 32(S1):128-133. (in Chinese)

王田, 刘伟宁, 韩广良, 等.基于改进MeanShift的目标跟踪算法[J].液晶与显示, 2012, 27(3):396-400.

WANG T, LIU W N, HAN G L, et al. Target tracking algorithm based on improved meanshift[J]. Chinese Journal of Liquid Crystals and Displays, 2012, 27(3):396-400. (in Chinese)

BVLTHOFF H, LITTLE J, POGGIO T. A parallel algorithm for real-time computation of optical flow[J]. Nature, 1989, 337(6207):549-553.

李英, 李静宇, 徐正平.结合SURF与聚类分析方法实现运动目标的快速跟踪[J].液晶与显示, 2011, 26(4):544-550.

LI Y, LI J Y, XU Z P. Moving fast target tracking using SURF and cluster analysis method[J]. Chinese Journal of Liquid Crystals and Displays, 2011, 26(4):544-550. (in Chinese)

张良国, 吴江琴, 高文, 等.基于Hausdorff距离的手势识别[J].中国图像图形学报, 2002, 7(11):1144-1150.

ZHANG L G, WU J Q, GAO W, et al. Hand gesture recognition based on Hausdorff distance[J]. Journal of Image and Graphics, 2002, 7(11):1144-1150. (in Chinese)

HU M K. Visual pattern recognition by moment invariants[J]. IRE Transactions on Information Theory, 1962, 8(2):179-187.

任彧, 顾程程.基于HOG特征和SVM的手势识别[J].科技通报, 2011, 27(2):211-214.

REN Y, GU C C. Hand gesture recognition based on HOG characters and SVM[J]. Bulletin of Science and Technology, 2011, 27(2):211-214. (in Chinese)

赵健, 张冬泉.基于OpenCV的数字手势识别算法[J].计算机应用, 2013, 33(S2):193-196.

ZHAO J, ZHANG D Q. Number gesture recognition method based on OpenCV[J]. Journal of Computer Applications, 2013, 33(S2):193-196. (in Chinese)

刘寅, 滕晓龙, 刘重庆.复杂背景下基于傅立叶描述子的手势识别[J].计算机仿真, 2005, 22(12):158-161.

LIU Y, TENG X L, LIU C Q. Hand gesture recognition based on Fourier descriptors with complex backgrounds[J]. Computer Simulation, 2005, 22(12):158-161. (in Chinese)

张嘉桐, 李雪妍, 郭树旭, 等.基于椭圆傅里叶描述子的形状表示的研究[J].计算机工程与应用, 2014, 50(2):170-174.

ZHANG J T, LI X Y, GUO S X, et al. Study on shape representation based on elliptic Fourier descriptor[J]. Computer Engineering and Applications, 2014, 50(2):170-174. (in Chinese)

GRANLUND G H. Fourier preprocessing for hand print character recognition[J]. IEEE Transactions on Computers, 1972, C-21(2):195-201.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

RGB深度图像显著性目标检测方法设计

扩散模型与傅里叶滤波器结合的图像去雾方法

基于颜色和光流的多注意力机制微表情识别

基于迭代传播网络的快速鲁棒的微光图像增强

视觉跟踪技术中孪生网络的研究进展