Contrastive_Learning_for_Blind_Hyperspectral_Unmixing.pdf (2.76 MB)
Download file

Contrastive Learning approach for blind Hyperspectral Unmixing (CLHU)

Download (2.76 MB)
posted on 2023-08-07, 13:15 authored by Abian Hernandez-GuedesAbian Hernandez-Guedes, Ryodai Fukushima, Toshihiro Takamatsu, Himar Fabelo, samuel ortega, Juan Ruiz-Alzola, Hiroshi Takemura, Gustavo M. Callico

The Contrastive Learning for blind Hyperspectral Unmixing (CLHU) is a self-supervised deep learning approach for blind hyperspectral unmixing. Unlike existing deep learning methods that rely on reconstruction capabilities, CLHU leverages the input-endmembers relationship for abundance estimation. The endmembers are dynamically updated during training, controlled by two regularization factors. Extensive experiments demonstrate CLHU's effectiveness, achieving state-of-the-art performance in hyperspectral unmixing. This novel approach offers a promising perspective for the field and holds potential for further enhancements in hyperspectral unmixing tasks.


“Agencia Canaria de Investigacion, Innovacion y Sociedad de la Información (ACIISI)” through Pre-Doctoral Grant by the “Consejería de Economía, Conocimiento y Empleo,”

European Union “NextGenerationEU/PRTR” (Grant Number: FJC2020-043474-I)

Japan Society for the Promotion of Science (Grant Number: JP21H03844)


Email Address of Submitting Author

ORCID of Submitting Author


Submitting Author's Institution

University of Las Palmas de Gran Canaria

Submitting Author's Country

  • Spain