多媒體資訊廣泛運用在網路中,尤其是彩色影像的應用更為廣泛,在網路頻寛及儲存空間的考量下,透過色彩量化的技術,將影像圖檔做一定比例的壓縮,在人眼可接受的影像品質損失下,顯示出與原圖差異甚小的圖像,除了節省儲存空間,也可以加速在網路上傳輸的速率。因此,如何使這些資訊能夠達到儲存最小化及圖像重現品質最佳化,色彩量化則是處理圖像壓縮的重要技術。 本論文提出一個新的彩色影像壓縮技術稱為ELBG,它改進LBG隨機選取編碼簿的方式,利用編碼簿大小決定RGB原色群集及原始影像像素集的分配,再過濾空值及重複像素以降低訓練像素的數目,取代LBG演算法以隨機方式所產生的初始編碼簿。實驗結果顯示,本論文所提出之ELBG演算法可提升壓縮品質及降低壓縮圖像時所耗費的時間。
In present, multimedia is a worldwide phenomenon in the internet, especially full-color image. An image conducts a certain percentage of compression by the color quantization scheme at an acceptable loss of quality to show little difference compared with original image. The color quantization is an important technique to compress these images for saving storage space and showing best image quality. This thesis proposes a color image compression technique called ELBG. It improves the design of codebook in the LBG by using codebook size to decide the training set, and reducing null values and duplication of training pixels of the number of pixels to replace the original LBG algorithm to generate a random way of the initial color palette for enhancing compression quality and time cost of image compression. Experimental results reveal that this algorithm outperforms LBG, H and F, SEAN and MLBG approaches.