A Methodology for Color Quality Improvement Using Image Segmentation

• Main Authors: Pin-Chau Peng, 彭品超
• Other Authors: Yui-Liang Chen
• Format: Others
• Language: en_US
• Published: 2005
• Online Access: http://ndltd.ncl.edu.tw/handle/83355415856275826122

碩士 === 世新大學 === 圖文傳播暨數位出版學研究所(含碩專班) === 93 === Color Management System (CMS) adjusts colors between different devices to achieve what you see is what you get. Color Management Modules (CMMs), components of CMS, are participated to contribute the better color quality. Not only the capability of CMM itself, but also the collaboration among CMMs, affects the quality of color reproduction. However, the imperfection of CMM, such as the roughness of colorimetric characterization algorithms, may limit the capability of CMM. Moreover, the lack of framework for collaborating different CMMs developed by different researchers may also the obstacle of color reproduction. The objective of this research is to provide approaches to improving color quality using K-means image segmentation algorithm and Color Management Web Services System (CMWSS). The system will described by Unified Modeling Language (UML) tool. In this research, the roughness of 3*3 matrix based color characterization method is improved by K-means segmentation. The suitable color spaces for K-means algorithm are found. Furthermore, using distributed CMS framework developed by web services and human acceptability factors to defined high quality color reproduction. The overall system describe by UML tool finally.