基于最大熵生长检测器的模糊红外图像分割算法  

Blurred infrared image segmentation based on maximum entropy growth detector algorithm

作　　者：于晓 吕欣欣 高强 叶溪 XIAO Yu;LV Xinxin;GAO Qiang;YE Xi(School of Electrical and Electronic Engineering, and Tianjin Key Laboratory for Control Theory & Applications in Complicated Systems, Tianjin University of Technology, Tianjin, 300384, China)

机构地区：[1]天津理工大学电气电子工程学院复杂系统控制理论与应用重点实验室,天津300384

出　　处：《激光杂志》2019年第3期68-73,共6页Laser Journal

基　　金：国家自然科学基金资助项目(NO.61502340);天津市自然科学基金资助项目(18JCQNJC01000);天津市教委科研计划项目(2018KJ133);2018天津理工大学重点实验室开放基金(TJKL-CTACS-201907)

摘　　要：针对图像背景复杂、目标边缘模糊的问题,提出了基于最大熵生长检测器的图像分割算法。根据图像的特点,首先利用多源区域生长法,定位目标,去除背景干扰及噪声;然后通过最大熵原理初步划分目标及背景区域,生成对应区域的图像;分别对这两类区域的图像进行特征训练,得到包含目标区域与背景区域信息的检测器;使用目标与背景的检测器对最大熵分割后的图像进行目标与背景的细分割,最终生成目标图像。实验结果表明,本算法具有良好的目标提取效果。In this paper,an image segmentation algorithm based on maximum entropy growth detector is proposed to solve the problem of image background complexity and target edge blur. Firstly,it uses the multi-source region growing method to locate the target according to the characteristics of the processed image,and removes the background interference and noise. The maximum entropy principle is utilized to roughly divide the target area and the background,and the image of the target and the background area is generated. Then we train the features of these two types of images and get the detector that contains the information of target area and background area. Using the detector of the target and the background,the image is segmented by the target and the background,and finally the target image is generated. Experimental results show that this algorithm has good effect on object extraction.

关 键 词：多源区域生长 最大熵检测器 图像分割 

分 类 号：TN911.4[电子电信—通信与信息系统]