基于显著性及主成分分析的红外小目标检测

将红外小目标检测作为目标与背景的二分类问题.先根据点扩散函数原理,仿真生成红外小目标训练样本,再用主成分分析方法提取目标样本的主特征,建立目标的主成分空间.对测试样本,只要判断其在主成分空间的重构残差,便可识别其是否为目标.为了提高算法的实时性,提出了一种基于显著性和主成分分析的红外小目标检测算法,先通过频域残差方法检测目标可能存在的显著性区域,再在此区域内做识别.实验结果证明该方法快速、有效.

INFRARED SMALL TARGET DETECTION BASED ON SALIENCY AND PRINCIPLE COMPONENT ANALYSIS

Infrared small target detection is considered as a binary classification problem between target and background. According to the principal point spread function (PSF), infrared small target training set was simulated. Principal component analysis (PCA) was used to extract the main characteristics of target sample.Thus, the principal component space of thr target was established. Each test sample can be recognized as either target or background by its reconstruction error in the principal subspace. In order to improve the real-time performance, an infrared small target detection algorithm based on saliency and PCA was proposed . Salient regions probably containing targets were firstly detected by using spectral residual approach. Then target recognition was performed in the salient regions. Experimental results indicate that the proposed algorithm is fast and effective.