Convolutional Autoencoder for Blind Hyperspectral Image Unmixing

TL;DR

This paper introduces a convolutional autoencoder architecture for blind hyperspectral image unmixing, effectively decomposing mixed pixels into endmembers and abundances with improved accuracy over existing methods.

Contribution

A novel convolutional autoencoder architecture designed specifically for blind hyperspectral unmixing, enhancing abundance estimation and endmember extraction performance.

Findings

Outperforms existing unmixing methods in abundance estimation

Achieves competitive endmember extraction results

Demonstrates effectiveness on real hyperspectral data

Abstract

In the remote sensing context spectral unmixing is a technique to decompose a mixed pixel into two fundamental representatives: endmembers and abundances. In this paper, a novel architecture is proposed to perform blind unmixing on hyperspectral images. The proposed architecture consists of convolutional layers followed by an autoencoder. The encoder transforms the feature space produced through convolutional layers to a latent space representation. Then, from these latent characteristics the decoder reconstructs the roll-out image of the monochrome image which is at the input of the architecture; and each single-band image is fed sequentially. Experimental results on real hyperspectral data concludes that the proposed algorithm outperforms existing unmixing methods at abundance estimation and generates competitive results for endmember extraction with RMSE and SAD as the metrics, respectively.