利用影像局部特徵的數位半色調法

大多數的顯示設備及印表機都是黑白的，所以也就無法輸出連續色調的影像，而數位半色調演算法是將數位灰階影像轉換成符合輸出設備特性的黑白點，使之造成包含許多不同種類的灰階效果；也就是說，若一個繪圖輸出裝置的每一像素只能顯示兩種強度水準，則可以使用半色調法提供影像中的強度變化。目前數位半色調演算法約可分為三類：誤差擴散法、次序過濾法及點擴散法。這三種研究方法各有優缺點，而且評估影像輸出品質也沒有客觀的標準方法，因此在這篇論文中，我們分別針對這三種方法，分析其擴散方法或過濾矩陣等關鍵性因素，提出改良的建議，我們是利用影像中的局部特徵，充份掌握影像中的重要訊息，在半色調處理過程中同時達到強化邊緣的效果，並減少所產生的花紋，我們所提出的三種方法，包括有灰階優先法、邊緣優先法和最接近優先法，最終的目的是在於提高輸出影像的品質。

關鍵字

數位半色調演算法；過濾矩陣；連續色調

並列摘要

Most printers and some display devices are bilevel (black or white) and therefore not capable of reproducing continuous tone pictures. Digital halftoning, also referred to as spatial dithering, is the method of rendering the illusion of continuous- tone pictures on displays that are capable of producing only binary picture elements. The most widely known and used digital halftoning algorithms are usually separated into three classes, (1) "error diffusion," (2) "ordered dither," and (3) "dot diffusion." The majority of existing halftoning algorithms were designed only a single template for general purpose. These algorithms perform poorly because the important information within the original image is absent. However, in any specific implementation of digital halftoning algorithms there is a trade-off between computational complexity and perceived image quality. This thesis describes some improvements for these three methods. The aim of this thesis is to take the image local feature and to propose three methods, called gray-priority method, boundary-priority method, and nearest-priority method, to halftone and enhance the output quality. It appears to avoid some deficiencies of other commonly used techniques. Various other artifacts associated with the halftoning process, such as contouring due to coarse quantization or to textual changes, are also reducible from the output images in the proposed scheme.

並列關鍵字

digital halftoning algorithm ； dither matrix ； continuous tone

國際替代計量

利用影像局部特徵的數位半色調法

主題瀏覽