关键词: 图像色域边界描述器, 线色域边界, 图像色域映射, 邻域插值

Abstract:

In view of the vacant-segment problems caused by the calculation of image gamut boundary using Segment Maxima Gamut Boundary Descriptor（SMGBD）, an interpolation algorithm for vacant segments using neighboring weighted average is proposed. Firstly, the gamut boundary point of the vacant-segment should be placed at the center of the segment to get the spherical coordinate angles. Then, by means of neighboring weighted average, the spherical coordinate radius of the segment is interpolated. To address the intersection problem caused by the coincidence between the line of two adjacent GBD points and the plane of constant hue angle, the intersection is set as the gamut boundary point of one of the gamut segment. In the assessment, the gamut-compressing evaluation based on psychophysical experiment is designed to verify the effect of the SMGBD interpolation algorithm. The result shows that the proposed interpolation algorithm can improve the precision of SMGBD and gamut mapping. In addition, the SMGBD interpolation has no influence on the computational efficiency of LGBs.

Key words: image gamut boundary descriptor, Line Gamut Boundary（LGB）, image gamut mapping, neighborhood interpolation