Graph Scaling Cut with L1-Norm for Classification of Hyperspectral Images

TL;DR

This paper introduces an L1-norm based graph method called L1-Scaling Cut for hyperspectral image classification, which is robust to noise and outliers and preserves data distribution.

Contribution

It proposes a novel L1-norm based dimensionality reduction technique for hyperspectral images, improving robustness over traditional L2-norm methods.

Findings

Effective on noisy hyperspectral data

Outperforms L2-norm based methods

Preserves data distribution

Abstract

In this paper, we propose an L1 normalized graph based dimensionality reduction method for Hyperspectral images, called as L1-Scaling Cut (L1-SC). The underlying idea of this method is to generate the optimal projection matrix by retaining the original distribution of the data. Though L2-norm is generally preferred for computation, it is sensitive to noise and outliers. However, L1-norm is robust to them. Therefore, we obtain the optimal projection matrix by maximizing the ratio of between-class dispersion to within-class dispersion using L1-norm. Furthermore, an iterative algorithm is described to solve the optimization problem. The experimental results of the HSI classification confirm the effectiveness of the proposed L1-SC method on both noisy and noiseless data.