基于双树复数小波和SVR的红外小目标检测  

Detection of small infrared target based on dual-tree complex wavelet transform and SVR

作　　者：吴一全[1,2] 尹丹艳[1] 纪守新[1] 

机构地区：[1]南京航空航天大学信息科学与技术学院,南京210016 [2]南京大学计算机软件新技术国家重点实验室,南京210093

出　　处：《仪器仪表学报》2010年第8期1834-1839,共6页Chinese Journal of Scientific Instrument

基　　金：国家自然科学基金(60872065);南京大学计算机软件新技术国家重点实验室开放基金(KFKT2010B17)资助项目

摘　　要：在分析红外图像弱小目标和背景特征的基础上,提出了基于双树复数小波变换(dual-tree complex wavelet transform,DT-CWT)和支持向量回归(support vectorr egression,SVR)的检测方法。首先采用双树复数小波变换抑制大部分背景噪声;其次用SVR对去噪后的红外图像进行背景预测,并用去噪后的实际图像减去预测图像得到残差图像,大大提高了图像的信噪比;接着提出了基于模糊Tsallis-Havrda-Charvat熵的阈值选取算法,对残差图像进行阈值分割;最后根据目标的连续性和运动轨迹的一致性检测出真实的小目标。实验结果表明:该方法可显著提高红外目标的检测概率,实现较远距离弱小目标的检测。Through analyzing the characteristics of small target and background in infrared images,a detection method based on dual-tree complex wavelet transform and support vector regression （SVR） is proposed.First,dual-tree complex wavelet transform is used to suppress most of the background noise.Then SVR is adopted to predict the background of the de-noised infrared image.The predicted image is subtracted from the de-noised source image,which gives a residual image.As a result,the signal-to-noise ratio （SNR） of the image is greatly improved.Then,a threshold selection method based on fuzzy Tsallis-Havrda-Charvat entropy is presented to segment the residual image.Finally,the small target is detected according to the target continuity and trajectory consistency.Experimental results show that the proposed method can significantly increase the detection probability of infrared target and achieve long-range small target detection.

关 键 词：红外弱小目标检测 双树复数小波变换 支持向量回归(SVR) Tsallis-Havrda-Charvat熵 

分 类 号：TN911.73[电子电信—通信与信息系统]