Fast Classification of Hyperspectral Images Using Globally Regularized Archetypal Representation With Approximate Solution
Abstract: Representation learning plays a crucial rule in pattern recognition. Recently, sparse representation (SR) has become a popular technique in high-dimensional high signal processing. In this paper, we provide an alternative archetypal representation (AR) to condu conduct ct the representation learning. Compared with SR, AR preserves the sparsity of the learned representation, but has a lower complexity and better interpretation. For the classification of hyperspectral image, spatial information plays an important role in iimproving mproving the classification performance. Instead of representing each sample individually, we propose a globally regularized AR model, which uses graph regularization to include the contextual information into the learned representation. Although the entire entir model is convex, it is time consuming to obtain the optimal solution, which limits its large-scale scale applications. To address the computational issue, we further propose an efficient approximate solver based on line search on a feasible solution trajectory.. It turns out that the approximate solution is very close the optimal solution, with a relative error less than 5% usually, but speeds up the optimization dozens of times. Experiments demonstrate that the learned representation with the approximate soluti solution on is discriminant comparable to the optimal solution. Using the learned representation as a high high-level level feature, a linear