复合顺序形态变换在红外图像边缘检测中的应用  

Infrared Image Edge Detection Based on Multiplex Order Morphology Transformation

作　　者：迟健男[1] 方帅[1] 曾鹏鑫[1] 徐心和[1] 

机构地区：[1]东北大学信息科学与工程学院,辽宁沈阳110004

出　　处：《东北大学学报（自然科学版）》2005年第2期103-106,共4页Journal of Northeastern University(Natural Science)

基　　金：国家高技术研究发展计划项目(2001AA22270)

摘　　要：针对红外图像对比度低、边缘模糊、噪声大、空间相关性强的特点,提出了一种基于复合顺序形态变换的红外图像边缘检测方法·通过图像的局部均值和方差自适应的调节增强系数进行图像增强;采用形态学边缘锐化算法改善边缘清晰度·根据复合顺序形态变换相关概念及性质构造3种广义形态学边缘检测算子,可有效抑制红外图像中的噪声,提取目标边缘·实验结果表明,这种算法可以有效地克服红外图像的缺陷,保持图像边缘细节,优于传统边缘检测器·A new method of infrared image edge detection based on multiplex order morphology transformation is put forward against such shortcomings of infrared image as low contrast, edge faintness, high noise level and spatial correlativity. The image is thus enhanced through adaptive accommodation coefficient determined by local mean value and variance; Arithmetic of morphology edge sharpening is used to improve edge definition. To extract edge and control noise of infrared image effectively, three generalized morphological operators are constructed on the basis of the concept and properties correlating to multiplex order morphology transformation. The experimental results showed that the method presented is more effective than conventional detection to overcome the defects of infrared image and keep fully on the subtle parts of image edges.

关 键 词：图像增强 边缘锐化 顺序形态变换 边缘检测 噪声抑制 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]