MiSiSUn: Minimum Simplex Semisupervised Unmixing

TL;DR

MiSiSUn introduces a novel semisupervised unmixing method that leverages data geometry with a simplex-volume penalty, significantly outperforming existing methods on simulated and real datasets.

Contribution

The paper presents the first integration of data geometry into library-based unmixing using a simplex-volume penalty within an archetypal analysis framework.

Findings

Outperforms state-of-the-art methods by 1-3 dB in simulated scenarios.

Achieves results close to geological maps on real datasets.

Provides open-source implementation and Python package for reproducibility.

Abstract

This paper proposes a semisupervised geometric unmixing approach called minimum simplex semisupervised unmixing (MiSiSUn). The geometry of the data was incorporated for the first time into library-based unmixing using a simplex-volume-flavored penalty based on an archetypal analysis-type linear model. The experimental results were performed on two simulated datasets considering different levels of mixing ratios and spatial instruction at varying input noise. MiSiSUn considerably outperforms state-of-the-art semisupervised unmixing methods. The improvements vary from 1 dB to over 3 dB in different scenarios. The proposed method was also applied to a real dataset where visual interpretation is close to the geological map. MiSiSUn was implemented using PyTorch, which is open-source and available at https://github.com/BehnoodRasti/MiSiSUn. Moreover, we provide a dedicated Python package for Semisupervised Unmixing, which is open-source and includes all the methods used in the experiments for the sake of reproducibility.