A New Color Reduction Scheme by K-Mean Clustering and Error-Diffusion for Improving Visual Quality
碩士 === 國立中興大學 === 資訊科學與工程學系 === 96 === The issue of color reduction is to represent the original graph by fewer colors. In this paper, k-mean clustering technique is applied to avoid color shifting problem, and additional weight factor is also applied on color replacing and diffusing to smooth the c...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2008
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/70667085140101631078 |
| Summary: | 碩士 === 國立中興大學 === 資訊科學與工程學系 === 96 === The issue of color reduction is to represent the original graph by fewer colors. In this paper, k-mean clustering technique is applied to avoid color shifting problem, and additional weight factor is also applied on color replacing and diffusing to smooth the color variations. The color reduced result of the proposed scheme has smooth color variations and does not have blocks of monotone color. The experiment shows that the proposed scheme has better visual results, and the comparison results of MPSNR and BPSNR metrics also conform to this improvement.
|
|---|