File Exchange

image thumbnail

HSI dictionary learning algorithm with SOMP sample selection

version 2.0.1 by Ayan Chatterjee
Sample Selection with SOMP for Robust Basis Recovery In Sparse Coding Dictionary Learning

Updated 18 Oct 2019

Abstract of the paper:
Sparse Coding Dictionary (SCD) learning is to decompose a given hyperspectral image into a linear combination of a few bases. In a natural scene, because there is an imbalance in the abundance of materials, the problem of learning a given material well is directly proportional to its abundance in the training scene. By a random selection of pixels to train a given dictionary, the probability of bases learning a given material is proportional to its distribution in the scene. We propose to use SOMP residue for sample selection with each iteration for a more robust or 'more complete' learning. Experiments show that the proposed method learns from both background and trace materials accurately with over 0.95 in Pearson correlation coefficient. Furthermore, the proposed implementation has resulted in considerable improvements in Target Detection with Adaptive Cosine Estimator (ACE).

Cite As

Chatterjee, Ayan, and Peter W. T. Yuen. “Sample Selection with SOMP for Robust Basis Recovery in Sparse Coding Dictionary Learning.” IEEE Letters of the Computer Society, vol. 2, no. 3, Institute of Electrical and Electronics Engineers (IEEE), Sept. 2019, pp. 28–31, doi:10.1109/locs.2019.2938446.

View more styles

Chatterjee, Ayan. Sample Selection with SOMP for Robust Basis Recovery In Sparse Coding Dictionary Learning. Code Ocean, 2019, doi:10.24433/CO.5073641.V2.

View more styles

Comments and Ratings (0)

Updates

2.0.1

doi and title updates

2.0.0

doi and title updates

MATLAB Release Compatibility
Created with R2019a
Compatible with any release
Platform Compatibility
Windows macOS Linux