Discriminative Robust Deep Dictionary Learning for Hyperspectral Image Classification

TL;DR

This paper introduces a robust deep dictionary learning framework for hyperspectral image classification, utilizing multiple dictionary levels and a discriminative final layer to improve accuracy against noisy data.

Contribution

A novel deep dictionary learning approach with a greedy training strategy and robustness to mixed noise, outperforming existing deep learning models on hyperspectral datasets.

Findings

Outperforms DBN, SAE, and CNN on benchmark datasets.

Robustness to Gaussian and sparse noise improves classification accuracy.

Greedy training of multiple dictionary levels enhances feature representation.

Abstract

This work proposes a new framework for deep learning that has been particularly tailored for hyperspectral image classification. We learn multiple levels of dictionaries in a robust fashion. The last layer is discriminative that learns a linear classifier. The training proceeds greedily, at a time a single level of dictionary is learnt and the coefficients used to train the next level. The coefficients from the final level are used for classification. Robustness is incorporated by minimizing the absolute deviations instead of the more popular Euclidean norm. The inbuilt robustness helps combat mixed noise (Gaussian and sparse) present in hyperspectral images. Results show that our proposed techniques outperforms all other deep learning methods Deep Belief Network (DBN), Stacked Autoencoder (SAE) and Convolutional Neural Network (CNN). The experiments have been carried out on benchmark hyperspectral imaging datasets.