A comprehensive review on sparse representation for image classification in remote sensing

The main objective of this paper is to provide a comprehensive study on Sparse Representation based feature extraction techniques in the image classification domain. Sparse Representation (SR) plays a vital role in both theoretical research and practical applications. The Sparse Representation being image dependent has become a broadly used feature extraction technique that represents the signal or image under study. Considering the feature extraction techniques, this review article includes the work involving Multikernel Fusion Sparse Representation. A successful classification of remote sensing data is a huge challenge because many factors belong to the appropriate selection of remote sensing data, image pre-processing and processing methods may result in improper results. This issue demands a detailed collection of information extraction algorithms which would be remarkably valuable for the researchers who are new to the field of high resolution remote sensing data in selecting a suitable classification technique. This paper concentrates on recapitulating all the possible information extraction techniques for image classification in remote sensing images.