基于背景抑制的改进Top-Hat红外小目标检测方法  

An Improved Top-Hat Infrared Small Target Detection Method Based on Background Suppression

作　　者：吴健 陆书文 芮大庆 缪欣 张俊举[1] WU Jian;LU Shu-wen;RUI Da-qing;MIAO Xin;ZHANG Jun-jul(Electronic Engineering and Opto-Electronic Engineering Institute,Nanjing University of Science and Technology,Nanjing 210094,China;Department of Automation,University of Science and Technology of China,Hefei 230027,China;Jiangsu Shuguang Opto-Electronics Co.,Ltd.,Yangzhou 225009,China)

机构地区：[1]南京理工大学电子工程与光电技术学院,南京210094 [2]中国科学技术大学自动化系,合肥230027 [3]江苏曙光光电有限公司,江苏扬州225009

出　　处：《电光与控制》2018年第9期42-44,共3页Electronics Optics & Control

摘　　要：针对红外小目标在背景较为复杂时边缘模糊不易被检测、虚警率高等问题,提出了一种基于SUSAN背景抑制的改进Top-Hat的检测方法。该方法先将原图像做SUSAN背景抑制,再对抑制后图像做开运算,最后将开运算结果与原图像做比较并分别赋值得到结果图像。为了验证改进Top-Hat方法的优越性,在图像做完背景抑制后分别做传统Top-Hat处理与改进后的Top-Hat处理,并对其结果进行了对比。结果表明,所提算法能够大大提高复杂背景下红外小目标的检测概率,有效地减少伪目标数量,即降低了虚警率。The infrared small targets in complex background are not easy to detect due to their fuzzy edges, and the false alarm rate is high. To solve the problems, an improved Top-Hat target detection method based on SUSAN background suppression is proposed. In this method, the background of the original image is suppressed by using SUSAN method, and then an opening operation is made to the suppressed image. Finally, the result of the opening operation is compared with the original image, and the result images are obtained by assigning values to them respectively. In order to verify the superiority of the improved Top-Hat method, the image after background suppression is processed respectively by using the traditional Top-Hat method and the improved Top-Hat method. The comparison results show that, the proposed algorithm can greatly improve the detection probability of infrared small targets in complex background, and effectively reduce the number of false targets, i. e., the false alarm rate.

关 键 词：红外探测 复杂背景 小目标检测 SUSAN背景抑制 TOP-HAT 

分 类 号：TN215[电子电信—物理电子学]