基于Zernike矩的低信噪比红外目标检测方法  

Target Detection Method for Low-Snr Infrared Images Based on Zernike Moments

作　　者：童卫红[1] 杜天军[1] 

机构地区：[1]西南技术物理研究所,四川成都610041

出　　处：《红外技术》2007年第10期607-611,共5页Infrared Technology

摘　　要：低信噪比红外图像普遍存在目标-背景间对比度较差、目标边缘模糊和噪声较大等特点,采用常规模板匹配、Hu矩方法难以取得理想检测效果。针对低信噪比红外目标图像,分析了Zernike矩的基本原理和计算方法,提出了基于Zernike正交矩的低信噪比红外目标检测方法,并比较了模板匹配、Hu矩、Zernike矩方法的目标识别效果。理论分析与实验验证了所提方法的有效性。Low-SNR infrared images often present bad feature such as low target-background contrast, blurred edge and large noise index, so traditional target recognition methods such as template matching and Hu moments cannot detect target effectively. Aimed at low-SNR infrared images, this paper analyzed principle of Zernike moment and it＇s computational methods, proposed a method for low-SNR infrared target detection based on Zernike moments, and compared target recognition effect among template matching, Hu moments and Zernike moments via experiment. Theoretical analysis and experiment result show that Zernike moments are superior to template matching and Hu moments in target detection, and the proposed method is effective for low-SNR infrared target detection.

关 键 词：低信噪比 红外图像 ZERNIKE矩 目标检测 

分 类 号：TN911.73[电子电信—通信与信息系统]