首页 | 本学科首页   官方微博 | 高级检索  
     

基于改进的SOM和ART的动态手势识别
引用本文:周维柏,李蓉. 基于改进的SOM和ART的动态手势识别[J]. 郑州大学学报(理学版), 2011, 43(2)
作者姓名:周维柏  李蓉
作者单位:1. 华南师范大学,增城学院教育信息技术部,广东,广州,511363
2. 华南师范大学增城学院计算机系,广东,广州,511363
摘    要:
提出一种动态手势识别算法,将动态手势识别问题转换为轨迹识别间题.首先以SOM算法作为分类器提取手势特征,将多维手势信息投影到二维平面中,根据每帧的顺序产生一平面轨迹,将产生的平面轨迹输人到改进的ART网络进行识别.实验结果表明,该算法用于动态手势识别是可行的且性能稳定.

关 键 词:自组织特征映射  自适应共振理论  动态手势识别

Dynamic Gesture Recognition Based on Improved SOM and ART Neural Network
ZHOU Wei-bai , LI Rong. Dynamic Gesture Recognition Based on Improved SOM and ART Neural Network[J]. Journal of Zhengzhou University(Natrual Science Edition), 2011, 43(2)
Authors:ZHOU Wei-bai    LI Rong
Affiliation:ZHOU Wei-bai1,LI Rong2 (1.Department of Education Information,Zengcheng College,South China Normal University,Guangzhou 511363,China,2.Department of Computer,China)
Abstract:
A dynamic gesture recognition algorithm was presented,which could convert the dynamic gesture recognition problem to track identification problem.First,the SOM algorithm as classifier was used to extract gesture characteristics,then the multi-dimensional gesture information was projected in the two-dimension plane.A plane trajectory was generated by the order of each frame,and the produced plane trajectory was input to identify by the improved ART network.The experiments showed that the algorithm for dynami...
Keywords:self-organizing map  adaptive resonance theory  dynamic gesture recognition  
本文献已被 CNKI 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号