影像從屬型色彩複製演算法之研究
No Thumbnail Available
Date
2005
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
本研究乃是藉由數學演算法，應用國際色彩組織 (International Color Consortium, ICC) 所訂定之標準色域壓縮方式，經過國際照明委員會 (Commission Internationale de l’clairage, CIE) 所製定之設備獨立色彩空間 (Device-independent Color Space, CIE Lab or CIE Luv) 進行色彩複製，在多媒體輸入輸出時，色彩能夠產生一致的效果。
本研究在三維色彩空間中進行色域壓縮時，將影像色彩資訊內容作為決定集中點之位置以及壓縮方式的主要依據。利用五種不同類型之影像 (花朵植物、人物、風景、高明度高彩度、極端影像) 做為實驗用之測試影像。分別以NC、NL、NNL、SC、SL、SNL六種不同方式之色彩複製演算法對每張影像進行處理，以印表機輸出之影像及螢幕原稿進行心理物理實驗。採用雙眼同時觀測視覺評比技術觀測影像，加上配對比較法進行實驗。最後將配對比較法之實驗數據統計整理之後，加以分析討論得到以下結果與發現：
一、藉由擴展影像之明度分佈，可以保持原始影像在暗部的階調層次。
二、色彩複製演算法模式之排名視影像內容而定。
三、花朵植物影像採用SL模式會得到最佳之色彩再現結果。
四、風景影像採用SL跟NL模式為最佳之方式。
五、滿版的色塊影像採用裁切法的方式來進行色域壓縮會得到最佳之效果。
六、極端影像採用NL跟SL模式為最佳之方式。
This study aims to achieve the consistency between colours in cross-media inputs and outputs in device-independent Color Space (CIE Lab or CIE Luv), stipulated by Commission Internationale de l’clairage (CIE), using algorithm and the standard gamut compression proposed by International Color Consortium (ICC). In order to establish a standardized image-dependent algorithm, the image-colour information contents are used to determine the focal point, the compression method in three-dimension gamut compression. Different types of images (flowers, figures, scenery, high brightness& chroma, and extreme images, were processed by different algorithms of color reproduction (NC, NL, NNL, SC, SL, SNL). Next, a psychophysical experiment was conduct, using Simultaneous-Binocular Viewing Technique and Pair Comparison Method to compare the images printed out by the algorithms with those shown on a screen. The statistic results shows that： 1.the tone scale in the low key of original images could be maintained by expanding the distribution of lightness. 2.the colors of flowers images was optimalized using SL algorithm. 3.as for scenery images, NL and SL algorithm were the best method for images processing. 4.if the images were in full tones, the gamut compression could be optimalized using clipping methods. finally, as for extreme images, NL and SL methods were the best methods for images processing.
This study aims to achieve the consistency between colours in cross-media inputs and outputs in device-independent Color Space (CIE Lab or CIE Luv), stipulated by Commission Internationale de l’clairage (CIE), using algorithm and the standard gamut compression proposed by International Color Consortium (ICC). In order to establish a standardized image-dependent algorithm, the image-colour information contents are used to determine the focal point, the compression method in three-dimension gamut compression. Different types of images (flowers, figures, scenery, high brightness& chroma, and extreme images, were processed by different algorithms of color reproduction (NC, NL, NNL, SC, SL, SNL). Next, a psychophysical experiment was conduct, using Simultaneous-Binocular Viewing Technique and Pair Comparison Method to compare the images printed out by the algorithms with those shown on a screen. The statistic results shows that： 1.the tone scale in the low key of original images could be maintained by expanding the distribution of lightness. 2.the colors of flowers images was optimalized using SL algorithm. 3.as for scenery images, NL and SL algorithm were the best method for images processing. 4.if the images were in full tones, the gamut compression could be optimalized using clipping methods. finally, as for extreme images, NL and SL methods were the best methods for images processing.
Description
Keywords
影像從屬, 色彩複製, 色彩管理, 色域對應, 獨立色彩空間