Spectral Image Repertoire dataset

Published: 11 October 2018| Version 1 | DOI: 10.17632/66sbcw6p29.1
Ignacio Lopez


This data is an spectral representation of a reduced set of images. The spectra was calculated by adding the values of each colour channel (R, G, B) and then finding its histogram. The added values represent a single colour where 0 is black and 765 (the maximum value) is white. The values in between correspond to the different colours expressed with 8-bits. The maximum number of bins in the histogram is 765. Each bins of the histogram correspond to an independent variable. The number of independent variables was reduced by reducing the number of bin to 256 bins when calculating the histogram.



Machine Learning