Gamut estimation with efficient sampling based on modified segment maxima

Ho-Gun Ha

doi:10.1016/j.displa.2017.02.001

Return

Gamut estimation with efficient sampling based on modified segment maxima

2017-07-01

PRE

Ho-Gun Ha

Shibudas Kattakkalil Subhashdas

Yeong‐Ho Ha *

DOI：10.1016/j.displa.2017.02.001

Original

Original request for help

Save

Abstract

En 中文

Gamut mapping is necessary to achieve color consistency between cross-media devices. In gamut mapping, accurate estimation of the gamut in each device is an important task because it directly influences on the quality of color consistency. However, depending on the samples or estimation method, a false gamut can be calculated, resulting in color distortion in the reproduced image. Accordingly, to address this problem, accurate gamut estimation with efficient sampling is proposed. The proposed method selectively determines the samples and plugs the local concavities formed from the segment maxima algorithm. We assumed that the surface of the RGB cube roughly corresponds to the surface of the real gamut. Thus, points on the surface of the RGB cube can be selected as samples. Furthermore, points around the primaries are more intensively selected than from other parts of the surface. The local concavities that generate a false gamut are plugged by using modified gamut boundary descriptors. A local concavity is detected using a CounterClockWise algorithm with three consecutive descriptors. The descriptor in a concavity region is then moved to a line connecting the preceding and subsequent descriptors. In experiments, the proposed method accurately estimates the gamut with a small number of samples when compared with previous methods, and largely reduces the color distortion in the reproduced images. (C) 2017 Elsevier B.V. All rights reserved.

Keywords:

Color management system

Device characterization

Gamut mapping

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.