Premium
Color‐vision theory and linear models of color vision
Author(s) -
Massof Robert W.
Publication year - 1985
Publication title -
color research and application
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.393
H-Index - 62
eISSN - 1520-6378
pISSN - 0361-2317
DOI - 10.1002/col.5080100302
Subject(s) - linear model , color vision , artificial intelligence , computer science , weighting , color constancy , linear system , mathematics , computer vision , set (abstract data type) , machine learning , image (mathematics) , mathematical analysis , medicine , radiology , programming language
This article presents an overview of linear color‐vision models and presents a review of a general vision theory that provides for the derivation of linear models from an initial set of basic assumptions. The general theory specifies that the input and output terms of linear color‐vision models are to be regarded as stimulus and sensory perturbations, respectively, rather than as absolute stimulus and sensory magnitudes. The general theory also specifies that the weighting coefficients of linear color‐vision models correspond to first‐order derivatives of some set of nonlinear functions that represent the state of the visual system. The general theory provides a unifying framework that connects linear color‐vision models with linear models for spatial and temporal aspects of vision.