首页 | 本学科首页   官方微博 | 高级检索  
     检索      

区域生长型分水岭算法及其在图像序列分割中的应用
引用本文:卢官明.区域生长型分水岭算法及其在图像序列分割中的应用[J].南京邮电大学学报(自然科学版),2000,20(3):51-54.
作者姓名:卢官明
作者单位:南京邮电学院信息工程系,江苏南京,210003
摘    要:针对物体基视频编码中的图像分割,提出了一种区域生长的分水岭算法,它直接对要分割的图像而不是其形态梯度信号进行分割,克服了传统形态分割方法使用形态梯度信号而导致轮廓信息丢失的缺点。此外,在具体实现算法时,采用了一种基于分级队列结构进行图像扫描的巧妙方法,大大提高了算法的效率。模拟结果表明,该算法能有效地分割运动物体,能较精确地定位运动物体的边缘。

关 键 词:图像分割  数学形态学  视频编码  分水岭
文章编号:1000-1972(2000)03-0051-04
修稿时间:1999-10-15

A Region-growing Version of Watershed Algorithm and Its Application to Segmentation of Image Sequences
LU Guan-ming.A Region-growing Version of Watershed Algorithm and Its Application to Segmentation of Image Sequences[J].Journal of Nanjing University of Posts and Telecommunications,2000,20(3):51-54.
Authors:LU Guan-ming
Institution:LU Guan-ming ;(Department of Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing 210003, China)
Abstract:This paper proposes a region growing version of watershed algorithm for image segmentation in the object based video coding system.This algorithm works directly on the image to be segmented in stead of the morphological gradient signal and solves the problem existing in the classical morplological segmentation approach in which the use of the morphological gradient results in a loss of contour information.Furthermore,in order to efficiently implement this algorithm,the paper proposes a smart image scanning approach based on hierarchical queues.Simulation results demonstrate the efficiency of the proposed algorithm.The main advantage of this algorithm is that the boundaries of segmented moving objects are precisely located.
Keywords:Image segmentation  Mathematical morphology  Video coding
本文献已被 CNKI 万方数据 等数据库收录!
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