Repository of hyperspectral images for image registration

  1. Ordóñez, Álvaro 1
  2. Blanco Heras, Dora 1
  3. Argüello, Francisco 1
  1. 1 Universidade de Santiago de Compostela
    info

    Universidade de Santiago de Compostela

    Santiago de Compostela, España

    ROR https://ror.org/030eybx10

Editor: Zenodo

Year of publication: 2021

Type: Dataset

License: Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International

DOI: 10.5281/ZENODO.10638886 lock_openOpen access editor

Abstract

This repository contains hyperspectral scenes used to test registration algorithms. The images are in two formats: MATLAB and RAW. A colour composition of these images, in PNG, is also available in the folders. In the RAW format, the hyperspectral data is stored as 32-bit signed integers. The pixels of each band are stored continuously (first band, second band...). A binary header of three integers (12 bytes) with the dimensions of the data (width, height, bands) is also included in the file. The images have to be extracted using 7z x file.raw.7z This format can be displayed with the linux program imagevis3d or with HypeRvieW. More information can be found in the README file and at https://gitlab.citius.usc.es/hiperespectral/RegistrationRepository