Bare-hand alphabets gesture recognition based on VCM and ROI segmentation

ZHOU Hang; RUAN Qiu-qi

您当前的位置：

首页 >

文章列表页 >

Bare-hand alphabets gesture recognition based on VCM and ROI segmentation

更新时间：2024-10-14

- Bare-hand alphabets gesture recognition based on VCM and ROI segmentation
- Issue 5, Pages: 94-101(2007)
- 作者机构：
  
  1. 北京交通大学信息科学研究所
  2. 北京交通大学信息科学研究所,北京,100044
- 作者简介：
- 基金信息：
- DOI：
  CLC： TP391.4
- Published：2007
- 稿件说明：
移动端阅览
ZHOU Hang, RUAN Qiu-qi. Bare-hand alphabets gesture recognition based on VCM and ROI segmentation[J]. 2007, (5): 94-101.
DOI：

ZHOU Hang, RUAN Qiu-qi. Bare-hand alphabets gesture recognition based on VCM and ROI segmentation[J]. 2007, (5): 94-101. DOI：

摘要

提出了实时条件下裸手单目的手部定位和跟踪算法。结合改进的差分法提出了实用的手部约束条件

从每帧图像中提取手部图像

确定感兴趣区域(ROI

region of interesting)

得到较为满意的手部分割结果

再对手部图像特征进行深入分析和提取

利用改进的相干映射算法(VCM

vector coherence mapping)进行跟踪

针对手的运动增加了约束

保证了顽健性。在这个基础上

提出了时间相关的运动预测模型

满足了实时性的要求

保证前后分析结果的一致性。实验结果证明

在不同光照和复杂背景下系统有最高达99%的识别率

与已有的系统相比

性能显著提高。

Abstract

A bare-hand gesture location and tracking system with single camera in real-time was introduced.By employing improved difference method

restrictions of hand are used to location region of Interesting（ROI）.Then on the basis of relatively satisfied segment result the thesis analyzed and extracted the hand features combined with the improved vector coherence mapping（VCM） whose constraint was enhanced to overcome the illegibility for robust results.Furthermore

a time-correlation forecasting model proved the demand of real-time and the consistency of the results now and before.The experiment results show the system has a high recognition rate up to 99% in different illuminations and backgrounds and has more efficient performance compared with existing other systems.

关键词

Keywords

references

Views

896

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

Review of HEVC video encryption technology

Research of dynamic gesture recognition based on multi-instance learning of kinematics features

Related Author

YE Qing

LIU Xiaobing

RONG Li

HE Junfei

ZHANG Qiaojia

Cai-qiu ZHOU

Yu-wang YANG

Hai-bo PANG

Related Institution

Department of Information Security, Naval University of Engineering

Base of Warship Testing and Training, Naval University of Engineering

School of Computer Science and Engineering，Nanjing University of Science and Technology

School of Software and Applied Science and Technology,Zhengzhou University

AI问答

⁰