基于改进分水岭算法的红外图像分割  

Infrared target segmentation method based on improved watershed algorithms

作　　者：孙伟[1] 王宏飞[1] 邵锡军[1] 

机构地区：[1]南京电子技术研究所,南京210013

出　　处：《红外与激光工程》2006年第z4期31-37,共7页Infrared and Laser Engineering

基　　金：国防预研基金(51407040105DZ2405)

摘　　要：研究自然背景下红外图像目标分割的问题,提出了一种基于改进分水岭算法的红外图像分割方法。该方法先利用形态学开闭滤波,对红外目标图像中的噪声和微小的干扰区域进行滤除,接着根据提出的计算图像形态梯度的多尺度算法提取图像梯度,而后用改进的分水岭算法对图像进行分割。最后提出了一种新的区域融合方法。实验结果表明,该算法能较好地解决红外图像中的目标分割问题。To segment targets under complicated background from infrared images of natural scenes,an infrared image segmentation algorithm based on watershed algorithms is presented in this paper.First,the noise and small interferential regions are removed with morphological filters.Second,a multiscale algorithm for computing morphological gradient images is proposed and the gradients are derived.Third,the images are segmented by improved watersheds and the problem of over-segmentation regions is settled well by an effective region merging method.The experiment results show the approach brought forth herein performs well in target segmentation in infrared images.

关 键 词：图像分割 形态梯度 分水岭算法 过分割 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]