飞机的红外图像分割算法  

A Segmentation Algorithm for Infrared Images of a Plane

作　　者：刘莞尔[1] 严高师[1] 戴霞[1] 

机构地区：[1]电子科技大学光电信息学院,四川成都610054

出　　处：《红外》2008年第3期27-31,共5页Infrared

摘　　要：本文针对飞机的红外图像的特点,提出了一种图像分割算法。首先采用整体与局部相结合的方法,在整体阈值的基础上,进行局部的适度调整,以达到最优阈值,使图像的分割在保持整体最优的基础上突出局部令人感兴趣的细节。然后利用数学形态学,针对目标内部的空洞区域进行填充。仿真结果表明,最后可以得到比较精确的分割图像。A novel image segmentation algorithm based on the features of the infrared images of a plane is proposed in this paper. First, the whole body is combined with its parts to make a proper adjustment for the parts on the basis of the integral threshold. Thus, the optimal threshold is obtained and the interested details of an object are emphasized while the best segmentation of the image is still kept. Then, the inside areas of the target image are filled by means of mathematical morphology. The simulation results show that a relatively accurate segmentation image can be obtained finally.

关 键 词：红外图像 阈值分割 形态学 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]