1. Convolutional neural network-based onboard band selection for hyperspectral data with coarse band-to-band alignment
- Author
-
Universitat Politècnica de Catalunya. Doctorat en Teoria del Senyal i Comunicacions, Universitat Politècnica de Catalunya. Departament de Teoria del Senyal i Comunicacions, Universitat Politècnica de Catalunya. CommSensLab-UPC - Centre Específic de Recerca en Comunicació i Detecció UPC, Llaveria Godoy, David, Longepe, Nicolas, Meoni, Gabriele, del Prete, Roberto, Camps Carmona, Adriano José, Universitat Politècnica de Catalunya. Doctorat en Teoria del Senyal i Comunicacions, Universitat Politècnica de Catalunya. Departament de Teoria del Senyal i Comunicacions, Universitat Politècnica de Catalunya. CommSensLab-UPC - Centre Específic de Recerca en Comunicació i Detecció UPC, Llaveria Godoy, David, Longepe, Nicolas, Meoni, Gabriele, del Prete, Roberto, and Camps Carmona, Adriano José
- Abstract
Band selection is a key strategy to address the challenges of managing large hyperspectral datasets and reduce the dimensionality problem associated with the simultaneous analysis of hundreds of spectral bands. However, the computational complexity of traditional methods makes the algorithms difficult to be deployed on board satellites. This is especially true for Small Satellites with limited computational and power resources. Moreover, existing band selection techniques often require the hypercube to be processed at least at Level-1B product, i.e., the bands need to be finely aligned before selecting them, demanding more computational resources for the on-board computer. This study presents a novel neural network-based approach for on-board band selection using data with coarse band-to-band aligned. This methodology not only simplifies the pre-processing requirements, but also opens new possibilities for efficient hyperspectral imaging from space on-board Small Satellites, such as classification, change and target detection., This project was part of the project "GENESIS: GNSS Environmental and Societal Missions – Subproject UPC", Grant PID2021-126436OB-C21 funded by the Ministerio de Ciencia e Investigación (MCIN)/Agencia Estatal de Investigación (AEI)/10.13039/501100011033 and EU FEDER “Una manera de hacer Europa”, and by a FPU fellowship from the Spanish Ministry of Education. Part of this work has also been possible thanks to the Italian Space Agency (ASI) that granted access to its PRISMA database (http://prisma.asi.it/)., Peer Reviewed, Postprint (published version)
- Published
- 2024