Hyperspectral small target detection by combining Kernel PCA with linear mixture model

In this paper, a kernel-based invariant detection method is proposed for small target detection of hyperspectral images. The method combines Kernel principal component analysis (KPCA) with linear mixture model (LMM) together. The LMM is used to describe each pixel in the hyperspectral images as a mixture of target, background and noise. The KPCA is used to build back-ground subspace. Finally, a generalized likelihood ratio test is used to detect whether each pixel in hyperspectral image includes target. The numerical experiments are performed on hyperspectral data with 126 bands collected by Airborne visible/infrared imaging spectrometer (AVIRJS). The experimental results show the effectiveness of the proposed method and prove that this method can commendably overcome spectral variability and sparsity of target in the hyperspectral target detection, and it has great ability to separate target from background.