[1]刘杨俊武,程春玲.基于关键帧和局部极值的手势特征提取算法[J].计算机技术与发展,2018,28(03):127-131.[doi:10.3969/ j. issn.1673-629X.2018.03.027]
　LIU Yang-junwu,CHENG Chun-ling.A Gesture Feature Extraction Algorithm Based on Key Frames and Local Extremum[J].,2018,28(03):127-131.[doi:10.3969/ j. issn.1673-629X.2018.03.027]
点击复制

基于关键帧和局部极值的手势特征提取算法()
分享到：

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
28
期数:
2018年03期
页码:
127-131
栏目:
智能、算法、系统工程
出版日期:
2018-03-10

文章信息/Info

Title:
A Gesture Feature Extraction Algorithm Based on Key Frames and Local Extremum
文章编号:
1673-629X(2018)03-0127-05
作者:
刘杨俊武; 程春玲
南京邮电大学 计算机学院,江苏 南京 210003
Author(s):
LIU Yang-junwu; CHENG Chun-ling
School of Computer,Nanjing University of Posts and Telecommunications,Nanjing 210003,China
关键词:
手势特征提取; 关键帧; 局部极值; 凸包
Keywords:
gesture feature extraction; key frames; local extremum; convex hull
分类号:
TP391
DOI:
10.3969/ j. issn.1673-629X.2018.03.027
文献标志码:
A
摘要:
为了缩短动态手势特征提取的时间,提出一种基于关键帧和局部极值的手势特征提取算法(KFLE)。 对传统关键帧图像选取算法进行改进,提出一种基于运动方向和自适应阈值的关键帧手势图像选取算法,通过比较相邻帧手势图像间的运动方向和质心距离,选择出运动方向或距离发生明显改变的关键帧手势图像进行指尖特征选取;在对关键帧手势图像进行特征提取的过程中,提出一种基于局部极值的指尖特征提取算法,通过构造距离函数并结合凸包过滤,寻找手势轮廓曲线上的局部极值点,确定手势中存在的指尖特征。 实验结果表明,与基于凸包缺陷的手势特征提取算法和基于改进 k-曲率的手势特征提取算法相比,该算法分别缩短了 44.3％和 71.9％的特征提取时间。
Abstract:
In order to shorten the time of dynamic gesture feature extraction,we propose a gesture extraction algorithm based on key frames and local extremum (KFLE). Firstly,we put forward an images selection algorithm of key frames based on motion direction and adaptive threshold for improvement of traditional images selection algorithm. By comparing the motion direction and centroid distance between adjacent frame gesture images,the images with significant changes in direction or distance are selected for feature extraction. In the process of feature extraction,we present a fingertip feature extraction algorithm based on local extremum. By constructing the distance function and combining the convex hull filtering,we find the local extreme points on the gesture contours to determine the characteristics of fingertip. The experiments show that the proposed algorithm can shorten the feature extraction time by 44. 3％ and 71.9％ respectively,compared with the gesture feature extraction algorithm based on convex hull defect and the gesture feature extraction algorithm based on improved k-curve.

相似文献/References:

[1]孙剑芬.基于高斯核密度估计的运动目标检测新方法[J].计算机技术与发展,2010,(08):45.
　SUN Jian-fen.New Method for Moving Object Detection Based on Gaussian Kernel Density Estimation[J].,2010,(03):45.
[2]赵一丹,肖秦琨,郭鹏.多目标优化的静态手语识别算法研究[J].计算机技术与发展,2019,29(02):54.[doi:10．3969/j．issn．1673－629X．2019．02．011]
　ZHAO Yidan,XIAO Qinkun,GUO Peng.Research on Static Sign Language Recognition Algorithm Based on Multi－objective Optimization[J].,2019,29(03):54.[doi:10．3969/j．issn．1673－629X．2019．02．011]

更新日期/Last Update: 2018-04-26