首页 | 本学科首页   官方微博 | 高级检索  
     

基于法向分量边缘融合的深度图像分割
引用本文:范剑英,于舒春,王洋,于贵江,于晓洋.基于法向分量边缘融合的深度图像分割[J].计算机工程,2010,36(17):221-222,225.
作者姓名:范剑英  于舒春  王洋  于贵江  于晓洋
作者单位:哈尔滨理工大学测控技术与通信工程学院,哈尔滨,150040
基金项目:国家自然科学基金,黑龙江省教育厅科学技术研究基金 
摘    要:针对深度图像边缘较难分割的问题,提出一种基于法向矢量分量边缘信息融合的深度图像分割方法。计算深度图像中每点的法向矢量,提取法向矢量的x、y分量并构建灰度图,融合Sobel算子的边缘检测结果得到深度数据的初始分割,通过细化处理得到最终分割结果。实验结果表明,该方法得到的分割区域边界闭合完整,分割质量较高。

关 键 词:深度图像分割  法向分量  边缘检测  信息融合

Range Image Segmentation Based on Edge Fusion of Normal Component
FAN Jian-ying,YU Shu-chun,WANG Yang,YU Gui-jiang,YU Xiao-yang.Range Image Segmentation Based on Edge Fusion of Normal Component[J].Computer Engineering,2010,36(17):221-222,225.
Authors:FAN Jian-ying  YU Shu-chun  WANG Yang  YU Gui-jiang  YU Xiao-yang
Affiliation:(College of Measurement-Control Technology & Communications Engineering, Harbin University of Science and Technology, Harbin 150040)
Abstract:Aimimg at the problem that range image is difficult to segment, this paper presents a range image segmentation method based on edge information fusion of normal components. Normal vector of every point in range image is computed, these normal vectors are decomposed into x component and y component which are used in building gray images. It uses Sobel operator in detecting edges of every range image and then these edges information are fused into an initial segmentation, and gets the final result by thin processing. Experimental results show that segmentation region edge closes completely and segmentation quality is above normal.
Keywords:range image segmentation  normal component  edge detection  information fusion
本文献已被 维普 万方数据 等数据库收录!
点击此处可从《计算机工程》浏览原始摘要信息
点击此处可从《计算机工程》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号