Services on Demand
Journal
Article
Indicators
Cited by SciELO
Access statistics
Related links
Similars in SciELO
Share
Computación y Sistemas
On-line version ISSN 2007-9737Print version ISSN 1405-5546
Abstract
DIAZ PAZ, Jean Pierre and LOAIZA CORREA, Humberto. Spherical Non-Perceptual Color Space RTP: Identifying Computer Graphics. Comp. y Sist. [online]. 2020, vol.24, n.1, pp.201-211. Epub Sep 27, 2021. ISSN 2007-9737. https://doi.org/10.13053/cys-24-1-3045.
This paper presents a non-perceptual color space based on spherical coordinates that is called RTP after Rho, Theta and Phi. Like HSV space color, RTP decouples the information associated to chrominance (θ,φ) and intensity (ρ), however RTP keeps together the reddish colors. To show the potentials of our color space, we approach the problem of identifying either an image is a computer graphic or a photorealistic image. We got an average accuracy of 84.55% detecting computer graphics and outperforming the ones from RGB, HSV and HSI color spaces with statistically significative differences on classification average accuracy. RTP color space can be used in other issues where chrominance decoupling improves the discriminant capability.
Keywords : Color space; computer graphics; image processing.