中國書法字型是世界最美且最富涵義的文字，各書法名家書寫於紙上或篆刻於碑文上的書法字型，經過時間歲月的?逝，往往會遺失或損壞，?人十分惋惜，故如何有效地保存固有傳統的書法字體，應是一值得探究的主題；本研究以對如何建?中國書法及碑刻的?位化保存為方向，以結合影像辨?處?的逆向設計技術，作為中國書法之?位化工程轉換依據，而使得中華傳統書法藝術得以永遠保存，並加以發揚光大。在書法字型逆向設計方法研究上，首先，藉由掃瞄器?取中文書法字型，並以灰預測??導入影像處?。在字型辨?方面，?用Matlab 軟體程式語言撰寫影像辨?程式，完成中國書法字型邊緣偵測；進而將字型影像邊緣??轉為點群資?，再將其匯入逆向CAD 軟體 ─Surfacer，以重建字型曲線??；再轉換成STL 格式，藉由快速原型機輸出字型實體，以達成字型之完整逆向設計與?位化的目標。在實?驗證上，本研究以中國書法中獨具特色的吉祥合成字型為案?，以毛筆書寫『招財進寶』一字，再經由掃描器將書寫於紙上之字型轉換為影像檔案，經由灰預測逆向辨?技術取得字型之影像邊緣，再將書法字型轉為點群文字檔，進?Spline 曲線建構，最後再?用快速原型，將其以3D ?體模型呈現，使中國書法字型能以?位化方式，並提供相關應用?域上展現出其獨特魅?與方寸之美。 Chinese calligraphy is thought as the most beautiful and meaningful patterns in the world. Calligraphy is an art dating back to the earliest day of Chinese history, and widely practiced throughout ancient China to these days. Owing to ages passing, many old carvings, stamping characters are gradually faded. How to reserve this old Chinese culture will be the crucial issues for today. On the other hand, the computer software for Chinese words processing can only deal with a few Chinese word styles, likes Ming, Kai, or Zhuan ...etc. There are still a lot of compound words or creating words used in different fields. This research is combined the reverse design, image processing and geometric modeling for keeping the characters of Chinese calligraphy. By this research, the gap between the Chinese calligraphy and computer technology can be compromised. This research developed a Chinese characters reverse design and pattern recognition procedure. The written Chinese calligraphy is scanned by scanner or pictured by CCD camera. By using grey prediction for image processing, the edges of the Chinese characters can be detected. The edges for the characters then turn out to be data points which can be easily processed be any CAD software by using B-Spline curves model. An example is demonstrated by writing a Chinese compounded word “the fortune and treasure are coming” to show the process. More usefully, with this research the Chinese characters can be reserved. Finally, rapid prototyping model is presented to show various applications for this research in Chinese calligraphy pattern recognition and processing.
