Please use this identifier to cite or link to this item: http://hdl.handle.net/11455/74778
標題: 新的具曲線筆觸繪圖成像演算法
作者: 王宗銘 
楊龍鴻
關鍵字: 圖像矩;曲線筆觸;繪圖成像;邊界測定
出版社: 國立中興大學工學院;Airiti Press Inc.
Project: 興大工程學刊, Volume 16, Issue 1, Page(s) 65-75.
摘要: 
Non-photorealistic rendering is a popular research topic in computer graphics communities, where painterly rendering has recently been under intensive studies. Painterly rendering algorithms based on the image moment have demonstrated some success, pro-ducing visually plausible painting images. However, most of these algorithms employed line-based brush strokes, disregarding the curved-stokes natures that have been adopted by painters in the real world. In this paper, we present a novel curved-strokes algorithm based on the image moment for painterly rendering. First, our algorithm employs the line integral convolution technique to determine a curved line passing thorough pixels coincident with the gradients of the image. Second, along with visiting every pixel in the line, we calculate the magnitude of the image moment to determine whether a stroke is to be drawn on the pixel currently visited, Next, the sizes and directions of the stroke are automatically determined we first compare the variances of the image moment at the current pixel, with those of the images moment in the neighborhood pixels. We then determine suitable magnitude of sizes and directions through linear interpolation calculation. Under this mechanism, a stroke with a small size and short length will be assigened to a pixel with small magnitude of the image moment. Third, we apply an image segmentation technique to identify boundaries of objects. When proceeding to painterly rendering, we can make sure that strokes will all be maintained inside the objects, without passing through the boundaries. Finally, we use a Gaussian filter on the original image and replace those pixels that is not drawn by any strokes. This allows us to simulate painterly rendering images when considering the existence of the canvas. We demonstrate some painterly rendering images, and compare our images with those proposed in the literature. Experimental results verify that our algorithm performs far better than the previous algorithms.

非擬真成像演算法為近期電腦圖學的熱門研究領域,繪圖成像更是其中重要的研究主題。結合圖像矩繪圖成像演算法已成功的產生具有類似畫家繪圖風格的圖像。然而,這些圖像矩演算法僅考慮直線的筆觸,無法繪製具有曲線筆觸的圖像。此不僅有違傳統畫家繪畫之特質,也使得所繪製的圖像顯得刻板僵硬。本文據此提出了一個全新的曲線圖像矩繪圖成像演算法。我們的演算法具有繪製曲線筆觸、多樣化的局部筆觸、維持圖像細節資訊等特性。就我們所知,這是個全新的圖像矩繪圖成像演算法,在文獻上屬於首創。首先,我們的演算法利用線積分迴旋,依據影像梯度並遵循其變化方向,模擬曲線筆觸的路徑。其次,我們計算圖像矩的標準法,以統計的技巧自動產生具有不同長短、大小、方向多樣變化的局部筆觸。再者,我們使用影像分割技巧來區分物體之邊界,讓繪製圖像時,筆觸不致橫跨邊界。最後,我們藉由對原始圖像高斯柔化技巧,將曲線筆觸未及繪製的像素點,化後的像素點取代,讓最後的結果更具有類似畫家繪圖風格的圖像。本文展示數張繪製的圖像,並與先前學者所產生的圖像做比較。結果顯示:我們的演算法所產生的圖像效果優於原始的直線圖像矩繪圖成像演算法。總結本文:我們提出一個全新的曲線圖像矩繪圖成像演算法,此演算法結合線積分迴旋、圖像矩標準差統計、影像分割、高質柔化等巧技。在繪製風格圖像時,新的演算法可遵循自動產生的筆觸長短、大小、方向,忠實模擬畫家的驗結果證實我們所提的方法優於先前學者的演算法。
URI: http://hdl.handle.net/11455/74778
ISSN: 1017-4397
Appears in Collections:第16卷 第1期
工學院

Show full item record
 
TAIR Related Article

Google ScholarTM

Check


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.