[1]刘 强,薛国新,薛 培.多分辨率下自适应阈值判定的分水岭车辆图像分割算法[J].常州大学学报(自然科学版),2011,(01):23-26.
　LIU Qiang,XUE Guo-xin,XUE Pei.Method of Adaptive Threshold Decision and Watersheds for Multiscale Image Segmentation[J].Journal of Changzhou University(Natural Science Edition),2011,(01):23-26.
点击复制

多分辨率下自适应阈值判定的分水岭车辆图像分割算法()
分享到：

常州大学学报(自然科学版)[ISSN:2095-0411/CN:32-1822/N]

卷:
期数:
2011年01期
页码:
23-26
栏目:
出版日期:
2011-01-01

文章信息/Info

Title:
Method of Adaptive Threshold Decision and Watersheds for Multiscale Image Segmentation
作者:
刘 强; 薛国新; 薛 培
常州大学 信息科学与工程学院,江苏 常州 213164
Author(s):
LIU Qiang; XUE Guo-xin; XUE Pei
School of Information Science and Engineering,Changzhou University,Changzhou213164,China
关键词:
图像分割;  自适应阈值判定;  分水岭算法
Keywords:
image segmentation;  adaptive threshold decision;  watershed
分类号:
TP 752.1
文献标志码:
A
摘要:
针对分水岭算法产生过分割现象,提出一种基于自适应阈值判定分水岭算法的图像分割方法,首先对图像进行多分辨率小波分解,再在某一尺度下计算梯度图像,并对该梯度图像进行自适应阈值判别,最后使用分水岭算法进行图像分割。实验结果表明该方法能取得较好的分割效果,有效的降低了过分割现象。
Abstract:
The problem of watershed transform is over-segmentation.This paper proposed a segmentation technique that combined adaptive threshold decision and watersheds for multiscale image segmentation. The wavelet transformthen was applied to the gray image, producing detail and approximation coefficients. Gradient magnitudes of the approximation image at the coarsest resolution were computed, and an adaptive threshold was used to remove small gradient magnitudes. The watershed transform was then applied. The result indicated that this method could generate a good segmentation for gray images and efficiently reduceover-segmentation.

参考文献/References:

[1]姜彬,施志刚.图像分割技术分析与展望[J].电脑知识与技术, 2009,26(35):31-35.
[2]Beucher S,Lantuejoul C. Use of Watersheds in Contour Detection[M]. Paris: International workshop for image Processing,1979:462-468.
[3]康牧,许庆功,王宝树.一种Roberts自适应边缘检测方法[J].西安交通大学学报,2008,26(5)27-32.
[4]袁春兰,熊宗龙,周雪花,等.基于Sobel算子的图像边缘检测研究[J]. 激光与红外,2009,26(1):44-48.
[5]杨道普,马秋禾,石磊.边缘检测Prewitt算子的改进算法[J].测绘科学,2008,25(33):26-30.
[6]Jong-Bae Kim,Hang-Joon Kim. Multiresolution-based watersheds for efficientimage segmentation[J]. Pattern Recognition Letters,2003,24:473-488.
[7]Scharcanski J,Jung C R,Clarke R T. A daptive image denosiing using scale and space consistency[J]. IEEE Transactions on Image Processing,2002,11(9):1092-1 101.
[8]吕建平,姬宝金.基于多分辨率分析与形态学分水岭算法的图像分割[J]. 微计算机信息, 2010,26(5):18-22.
[9]王国权,周小红,蔚立磊.基于分水岭算法的图像分割方法研究[J]. 计算机仿真, 2009,26(5):39-43.

相似文献/References:

[1]陈树越,邹　凌.基于光纤锥成像的织物疵点检测[J].常州大学学报(自然科学版),2009,(01):62.
　CHEN Shu -y ue,ZO U Ling.Research on Inspection of Fabric Surface Based on Optical Fiber Taper[J].Journal of Changzhou University(Natural Science Edition),2009,(01):62.

备注/Memo

备注/Memo:
作者简介:刘强(1979-),男,江苏南京人,硕士生。
更新日期/Last Update: 2011-01-01