Sparse representation based multi-threshold segmentation for hyperspectral target detection

Abstract

A sparse representation based multi-threshold segmentation (SRMTS) algorithm for target detection in hyperspectral images is proposed. Benefiting fro m the sparse representation, the high -dimensional spectral data can be characterized into a series of sparse feature vectors which has only a few nonzero coefficients. Through setting an appropriate threshold, the noise removed sparse spectral vectors are divided into two subspaces in the sparse domain consistent with the sample spectrum to separate the target from the background. Then a correlation and a vector 1-norm are calculated respectively in the subspaces. The sparse characteristic of the target is used to ext ract the target with a mu lti -threshold method. Un like the conventional hyperspectral dimensionality reduction methods used in target detection algorithms, like Principal Co mponents Analysis (PCA) and Maximu m No ise Fraction (MNF), this algorithm maintains the spectral characteristics while remov ing the noise due to the sparse representation. In the experiments, an orthogonal wavelet sparse base is used to sparse the spectral information and a best contraction threshold to remove the hyperspectral image noise according to the noise estimation of the test images. Co mpared with co mmon algorith ms, such as Adaptive Cosine Estimator (A CE), Constrained Energy Minimizat ion (CEM ) and the noise removed MNF-CEM algorith m, the proposed algorithm demonstrates higher detection rates and robustness via the ROC curves. INRODUCTIONRecently, hyperspectral imag ing detection technology is widely used in many applications , especially in the military.Providing with abundant spectral information, however, hyperspectral imaging has some challenges in processing the redundancy of spectral informat ion and the calculation co mp lexity of h igh-dimensional data due to an increase of the number of spectral bands. Moreover, there always exist some noises derived fro m the sensor itself and the erro r