北京大学学报(自然科学版) ›› 2015, Vol. 51 ›› Issue (6): 999-1007.DOI: 10.13209/j.0479-8023.2015.059

上一篇    下一篇

一种快速高效的手势跟踪识别方法

全冬兵1,2, 程如中2, 赵勇2, 魏江月2, 梁浩2, 魏益群3   

  1. 1. 中国工程物理研究院计算机应用研究所, 绵阳 621900
    2. 北京大学信息科学技术学院, 北京 100084
    3. 深港产学研基地, 深圳 518057
  • 收稿日期:2014-09-28 出版日期:2015-11-20 发布日期:2015-11-20
  • 通讯作者: 赵勇, E-mail: yongzhao(at)pkusz.edu.cn
  • 基金资助:
    中国工程物理研究院科学技术发展基金(2012A0403021)和深圳技术创新计划(CXZZ20120831104503786)资助

A Real Time and Effective Method for Hand-Gesture Detection and Tracking

QUAN Dongbing 1,2, CHENG Ruzhong 2, ZHAO Yong 2, WEI Jiangyue 2, LIANG Hao 2, WEI Yiqun 3   

  1. 1. Institute of Computer Application, China Academy of Engineering Physics, Mianyang 621900
    2. School of Electronics Engineering and Computer Science, Peking University, Beijing 100084
    3. PKU-HKUST ShenZhen-HongKong Institution, Shenzhen 518057
  • Received:2014-09-28 Online:2015-11-20 Published:2015-11-20
  • Contact: ZHAO Yong, E-mail: yongzhao(at)pkusz.edu.cn

摘要:

为了降低识别复杂度, 提高识别效率, 实现手势的快速高效跟踪, 提出一种分情况检测思想和搜索框概念。首先对图像进行细检测, 得到目标的准确位置, 然后通过粗检测与跟踪相结合的方式进行目标跟踪, 并对跟踪结果进行修正和可信度判断。实验结果显示: 算法对图像手势的平均检测跟踪正确率可以达到
97.36%, 且保证平均漏检率在5%以下, 对各种外界因素具有较好的鲁棒性; 算法对视频图像的处理速度达19.42 帧/秒, 满足人机交互系统中的实时性需求; 与TLD 算法相比, 本算法在处理速度上有数量级的改善, 算法结果的正确率也有明显优势。

关键词: 目标跟踪, 人机交互, 分情况检测, 搜索框

Abstract:

An algorithm for gesture detection and tracking in HCI (human-computer interaction) is designed to meet real-time and accuracy requirements. An innovative conception, which includes using distinguishing detection methods to detect hand-gesture for different conditions and using searching-box to decrease the searching zone, is proposed. The result shows that the detection rate can reach 97.36% while the missing rate lowers than 5%. It is robust to various external factors. However, it also meets the real-time, as the frame rate can reach 19.42. Compared with TLD, this algorithm has not only magnitude improvement in processing speed but also obvious advantages in accuracy.

Key words: object tracking, human-machine interaction, distinguishing detection, searching box

中图分类号: