A new bandwidth selection criterion for using SVDD to analyze hyperspectral data

TL;DR

This paper introduces an automatic, unsupervised method for selecting the Gaussian kernel bandwidth in SVDD, improving hyperspectral image classification accuracy across multiple datasets.

Contribution

It proposes a novel unsupervised bandwidth selection technique for SVDD, enhancing multiclass hyperspectral data classification performance.

Findings

Better classification accuracy than previous methods

Effective on multiple hyperspectral datasets

Automates kernel bandwidth selection process

Abstract

This paper presents a method for hyperspectral image classification that uses support vector data description (SVDD) with the Gaussian kernel function. SVDD has been a popular machine learning technique for single-class classification, but selecting the proper Gaussian kernel bandwidth to achieve the best classification performance is always a challenging problem. This paper proposes a new automatic, unsupervised Gaussian kernel bandwidth selection approach which is used with a multiclass SVDD classification scheme. The performance of the multiclass SVDD classification scheme is evaluated on three frequently used hyperspectral data sets, and preliminary results show that the proposed method can achieve better performance than published results on these data sets.