欢迎访问《图学学报》
分享到：
摘 要：根据手指几何形状大致平行的特性，本文提出一种基于曲率和平行向量#br# 来进行手指检测的新方法。该方法首先利用深度图像信息从复杂背景环境中迅速分割出手#br# 部；然后依据形态学操作和中值滤波进行平滑处理后提取边缘，再根据曲率定位拟指尖点；#br# 最后采用平行向量特性排除误检点。实验结果表明，该方法在不同手指运动状态，不同光照#br# 强度，以及复杂环境背景下均能实时定位指尖位置，指尖位置识别率可达98.64%。#br# 关 键 词：指尖检测；深度图像；曲率；平行向量
Abstract: According to the feature of the fingers' geometry being roughly parallel, this paper#br# proposes a new method based on curvature and paralleled-vector for fingertip detection. This#br# method first uses depth image information from complex background environment to segment#br# hand quickly, and then based on morphological operations and median filter for smoothing, edges#br# can be extracted. Subsequently, we can use curvature to locate the position of the virtual#br# fingertips. Finally the paralleled-vector feature is adopted to rule out error detection points. The#br# experimental results show that the method in different finger motions, different light intensities,#br# and complex environment can locate the position of the fingertips. Besides, the fingertips#br# detection rate can reach 98.64%.#br# Key words: fingertip detection; depth image; curvature; paralleled-vector
摘要： 根据手指几何形状大致平行的特性，本文提出一种基于曲率和平行向量
来进行手指检测的新方法。该方法首先利用深度图像信息从复杂背景环境中迅速分割出手
部；然后依据形态学操作和中值滤波进行平滑处理后提取边缘，再根据曲率定位拟指尖点；
最后采用平行向量特性排除误检点。实验结果表明，该方法在不同手指运动状态，不同光照
强度，以及复杂环境背景下均能实时定位指尖位置，指尖位置识别率可达98.64%。
