Losslessly Index Compression for Palette Images

• Main Authors: Sing-Hua Hsiao, 蕭幸樺
• Other Authors: Yu-Chin Hu
• Format: Others
• Language: zh-TW
• Published: 2010
• Online Access: http://ndltd.ncl.edu.tw/handle/15776597153479228811

碩士 === 靜宜大學 === 資訊管理學系研究所 === 98 === This paper presents a new index compression method for indexed color images. The proposed method can be viewed as the losslessly post processing of the index table generated by color image quantization. All the indices to be processed can be classified into three categories and three different rules are used to encode them. In the proposed method, the indicators of these three categories are dynamically assigned based on the occurrences of them. From the results, it is shown that the proposed method significantly reduces the required bit rates while consuming a low computational cost.