基于NMF、ICA和复Contourlet变换的红外小目标检测

摘要

Aimed at the detection problem of dim target in infrared image in the present of background interference and noise, a detection method based on non-negative matrix factorization( NMF) , independent component analysis( ICA) and complex Contourlet transform is proposed. The background of original image is suppressed respectively by using non-negative matrix factorization and independent component analysis, and different residual images of small target are obtained. The residual images are denoised by using complex Contourlet transform. Then addition of the above-mentioned denoised residual images gives a preprocessed image. At last, the preprocessed image is segmented by using the threshold segmentation based on fuzzy gray entropy, so that the dim target is detected under complex background. Lots of experiments are done for infrared images with small targets, and a comparison is made by using detection methods based on new Top-hat transform and fast independent component analysis. The experimental results show that the suggested method is stronger in anti-noise performance, and more superior in detection performance.%针对存在背景干扰和噪声情况下的红外弱小目标检测问题,提出了一种基于非负矩阵分解(NMF)、独立分量分析(ICA)和复Contourlet变换的检测方法.首先通过非负矩阵分解和独立分量分析分别抑制原始图像的背景,得到不同的小目标残差图像；接着采用复Contourlet变换对残差图像进行去噪；再对上述去噪后的小目标残差图像求和,得到了预处理图像；最后提出基于模糊灰度熵阈值选取方法分割预处理图像,从而实现了复杂背景下的红外弱小目标检测.针对红外小目标图像进行了大量实验,并与基于新型Top-hat变换、基于快速独立分量分析的目标检测方法进行了比较,结果表明所提出的方法抗噪性强,具有更为优越的检测性能.