现代电子技术2024，Vol.47Issue(15)：81-85,5.DOI:10.16652/j.issn.1004-373x.2024.15.013
基于改进凸包算法的手势定位
Gesture positioning based on improved convex hull algorithm
郭金朋 1韩昌林 2陆洲 2邵阳3
作者信息
- 1. 浙江宇视科技有限公司,浙江 杭州 310051
- 2. 安徽开放大学,安徽 合肥 230022
- 3. 广东工业大学,广东 广州 510006
- 折叠
摘要
Abstract
In this paper,the gesture segmentation,gesture contour,gesture convex hull,convex hull defect,fingertip positioning and gesture tracking are introduced mainly,and the simulation of keyboard and mouse is realized based on this architecture.The image is transformed from RGB model to HSV(hue,saturation,value)model,and then the skin color threshold is used to segment the hand according to HSV,so as to obtain the convex hull of gesture contour.In this paper,the number of the fingertips and the convex hull defects is calculated by principle of triangle,which solves the problem that convex hull defects fail to be positioned in the horizontal state of gesture.The gesture tracking algorithm with monocular vision is used to perform fingertip positioning initialization object tracking in order to realize unsupervised Camshift real-time gesture tracking.The experimental results show that the accuracy rate of keyboard simulation is over 90%under the conditions of good lighting.关键词
指尖定位/手势跟踪/键鼠模拟/凸包缺陷/单目视觉/帧差法/色彩模型/图像增强Key words
fingertip positioning/gesture tracking/simulation of keyboard and mouse/convex hull defect/monocular vision/frame difference method/color model/image enhancement分类
信息技术与安全科学引用本文复制引用
郭金朋,韩昌林,陆洲,邵阳..基于改进凸包算法的手势定位[J].现代电子技术,2024,47(15):81-85,5.
