基于三维块匹配与改进Top-hat的红外图像目标检测方法  

3-D Block-Matching Filtering and Improved Top-hat Method for Infrared Image Target Detection

作　　者：马烜 邹金慧[1,2] MA Xuan;ZOU Jinhui(Faculty of Information Engineering and Automation,Kunming University of Science and Technology,Kunming 650500,China;Engineering Research Center for Mineral Pipeline Transportation of Yunnan Province,Kunming 650500,China)

机构地区：[1]昆明理工大学信息工程与自动化学院,云南昆明650500 [2]云南省矿物管道输送工程技术研究中心,云南昆明650500

出　　处：《探测与控制学报》2019年第6期83-87,94,共6页Journal of Detection & Control

基　　金：国家自然科学基金项目资助(61663017)

摘　　要：针对复杂背景干扰下难以精确检测出红外图像目标的问题,提出基于三维块匹配(BM3D)与改进顶帽(Top-hat)的红外图像目标检测方法。该方法首先采用三维块匹配算法对红外图像进行滤波,更好地保留图像的边缘信息;其次构建改进Top-hat算子,利用不同大小、不同形状的结构元素对滤波后图像进行背景估计,得到校正后图像;最后对校正后图像进行阈值分割,得到目标图像。仿真实验结果表明,与经典Top-hat算法比较,提出的方法能够有效地增强红外图像对比度、抑制噪声干扰、减弱非均匀加热背景的影响,从而突出红外图像目标信息,使得红外图像目标检测更加准确。In view of the difficulty in accurately detecting infrared image targets under complex background interference,this paper proposed a Block-Matching and 3-D Filtering(BM3D)and improved top-hat infrared image target detection method.The method firstly used the Block-Matching and 3-D to filter the infrared image to better preserve the edge information of the image.Secondly,the improved Top-hat operator was constructed.The background image of the filtered image was estimated by using structural elements of different sizes and shapes,corrected image.Finally,the corrected image was subjected to threshold segmentation to obtain a target image.The simulation results showed that compared with the classical Top-hat algorithm,the proposed method could effectively enhance the infrared image contrast,suppress noise interference,and weaken the influence of non-uniform heating background,thus highlighting the infrared image target information and making the infrared image target detection more accurate.

关 键 词：三维块匹配 改进顶帽 阈值分割 无损检测 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]