彩色電子紙的高效前光選擇法

Abstract

彩色電子紙透過黑白電子泳顯示器搭配彩色濾光片呈現色彩，但受限於結構，亮度與色彩飽和度不足。前光面板可提升亮度與色彩，然而不同前光組合對色彩再現影響顯著，如何依影像內容選擇最適前光成為關鍵課題。本研究提出根據影像內容決定前光選擇演算法，依據輸入影像的代表色，自動推估最佳前光參數。方法為先分析影像主色調，並利用反距離加權法，從 27 組量測樣本中內插最佳前光組合，再以三維三角剖分模型估算該前光下的色域範圍。為修正前光引起的整體色偏，建立色彩轉換矩陣將影像由標準光源色域轉換至當前前光條件，並計算色彩校正矩陣補償裝置非線性失真。運算過程以粒子群最佳化加速，同時結合亮度增強與誤差擴散半色調技術，提升細節與低亮區域可視性，並符合彩色電子紙顯示特性。結果顯示，本方法能有效改善不同前光條件下的色彩一致性與影像品質，具實用性與可行性。

Color electronic paper (e-paper) uses a monochrome electrophoretic display with a color filter array, but structural limits reduce brightness and color saturation. A front light panel can improve both, yet different settings significantly affect color reproduction. This study proposes an image-content-based algorithm that selects optimal front light parameters from dominant image colors. Using inverse distance weighting over 27 measured samples and 3D Delaunay triangulation, it interpolates parameters and estimates the gamut. A color transformation matrix corrects lighting-induced shifts, while a correction matrix compensates for device nonlinearities. Particle swarm optimization accelerates computation, and brightness enhancement with error diffusion halftoning improves detail and low-light visibility. Results show improved color consistency and image quality under various lighting, confirming the method’s practicality and feasibility.