首页    编委会    编辑部    历任主编    获奖信息    征稿简则    友情链接    English    旧版主页
  在线期刊
       最新录用
       当期目次
       过刊浏览

  下载专区
       版权协议
       征稿范文
       来稿登记表
       录用证明

  联系我们

投稿e-mail: jbnuns_sub@bnu.edu.cn

主编e-mail: jbnuns_eic@bnu.edu.cn

办公地点:北京师范大学 科技楼A区216,218,220室

电话:(010)58807851,58802050,58807872,58807713

通信处:北京市海淀区新街口外大街19号北京师范大学理科学报编辑部

邮政编码:100875

办公室e-mail: jbnuns@bnu.edu.cn

本网站暂不提供全文下载,如需全文下载请点击

中国知网

 

标题

利用Kinect手部跟踪构造手势空间点集进行指尖定位的研究与实现

作者

杨博雄 胡旭华 黄 静

机构

北京师范大学珠海分校信息技术学院; 珠海市图形图像公共实验室

摘要

利用Kinect实现非接触式人机交互是当前研究的热点,手势识别是Kinect应用程序的核心,但是Microsoft KinectforWindowsSDK提供的手部深度图像跟踪算法中没有手指的描述子集,因此不能进行有效的指尖定位.基于 此,本文提出利用Kinect手部跟踪点获取真实坐标,进而构造空间点集来提取手势信息,其中空间跟踪点集是指满足目 标特征且与局部跟踪信息相关的空间坐标集合.在此基础上,利用K-曲率算法实现手势的指尖定位,并通过9个经典手 势对该方法进行验证.实验结果表明,通过本文提出的实现方法所获取的指尖位置准确率可达到91%,指尖定位速度可 以达到30FPS,完全可以满足体感游戏、页面自动换页等功能的需要.

关键词

Kinect;手部跟踪;空间跟踪点集;K-曲率算法;指尖定位

引用

杨博雄, 胡旭华, 黄 静. 利用Kinect手部跟踪构造手势空间点集进行指尖定位的研究与实现[J]. 北京师范大学学报(自然科学版),2015,51(6):571-575.

基金

国家自然科学基金资助项目(61272364)

分类号

TP391.4

DOI

10.16360/j.cnki.jbnuns.2015.06.006

Title

Fingertip Location Through the Gesture Space Points Sets Constructed by Kinect Hand Tracking

Author

YANG Boxiong, HU Xuhua, HUANG Jin

Affiliations

Information Technology College, Beijing Normal University;Public Image and Graphics Laboratory of Zhuhai City

Abstract

Non-contact human-computer interaction was realized using Kinect. Gesture recognition is the core of Kinect applications. Effective finger positioning cannot be obtained because image tracking algorithm for hand depth provided by Microsoft Kinect for Windows SDK has no finger descriptor sets. This paper therefore proposed a method to obtain real coordinates by Kinect hand tracking point and by extracting gesture information constructing space point sets, the space tracking point sets is the spatial coordinates sets that meet the target characteristics and is associated with the local track information. Fingertip location is therefore implemented through K-curvature algorithm and verified by nine classic gestures. It was found that fingertip position accuracy reached 91% and the finger positioning speed reached 30 FPS proposed by the realization method in this paper, which could fully meet the needs of function such as somatosensory games, and automatic paging.

Key words

kinect; handle Tracking; space tracking point sets; K-curvature algorithm; fingertips location

cite

YANG Boxiong, HU Xuhua, HUANG Jin.Fingertip Location Through the Gesture Space Points Sets Constructed by Kinect Hand Tracking[J]. Journal of Beijing Normal University(Natural Science),2015,51(6):571-575.

DOI

10.16360/j.cnki.jbnuns.2015.06.006

Copyright © 2014 Journal of Beijing Normal University (Natural Science)
Designed by Mr. Sun Chumin. Email: cmsun@mail.bnu.edu.cn