Contour Enhancement Algorithm for Improving Visual Perception of Deutan and Protan Dichromats

• Autor(a) principal: Ribeiro, Madalena
• Data de Publicação: 2019
• Outros Autores: Gomes, Abel
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Científico de Acesso Aberto de Portugal (Repositórios Cientìficos)
• Texto Completo: http://hdl.handle.net/10400.6/8293
• topic: Recoloring algorithm; Colorblindness

A variety of recoloring methods has been proposed in the literature to remedy the problem of confusing red-green colors faced by dichromat people (as well by other color-blinded people). The common strategy to mitigate this problem is to remap colors to other colors. But, it is clear this does not guarantee neither the necessary contrast to distinguish the elements of an image, nor the naturalness of colors learnt from past experience of each individual. In other words, the individual’s perceptual learning may not hold under color remapping. With this in mind, we introduce the first algorithm primarily focused on the enhancement of object contours in still images, instead of recoloring the pixels of the regions bounded by such contours. This is particularly adequate to increase contrast in images where we find adjacent regions that are color-indistinguishable from the dichromacy’s point of view.