期刊文献+

基于唐卡图像的线条画提取方法研究 被引量:2

Study of Line Drawing Extraction Method Based on Thangka Image
下载PDF
导出
摘要 唐卡以线条为骨架造型,是色彩和线条有机组合。唐卡图像的线条画提取旨在提取一组连贯、光滑、及具风格的线条,有效地捕获和传达唐卡的轮廓信息。以有限的信息量展现唐卡的内容,赋予唐卡另外一种艺术展现形式。本文针对唐卡图像特点提出其线条画提取方法。首先,根据唐卡色彩鲜明的特点,提取RGB颜色空间三个通道梯度向量并进行融合;然后,用边缘切向流方法构造一个光滑的方向场,保留了突出的图像特征。最后,通过基于流的高斯差分滤波方法提取连贯一致的线条,同时有效地抑制噪声干扰。实验证明该方法简单,容易实现,能有效提取唐卡图像线条。 Thangka of Thangka is the organic combination of color and lines ,whose skeleton model are lines. Line drawing image is to extract a set of lines which are coherent, smooth, and stylized, capture and convey the contour information of Thangka effectively. With the purpose of to express thangka's content using limited amount of information and exhibit it using another style of art. The paper presents a non-photorealistic rendering technique for Thangka image in the line drawing style. Firstly, according to the color features of Thangka image the gradient vector in RGB three color channel are extracted and fused. Then, a smooth direction field using edge tangent flow are constructed , which can preserve the salient feature of image. Finally, through the flow-based difference-of-gaussians filter the coherent lines are extracted . In addtion, the noise is supressed in the paper. Experimental results demonstrate that the proposed method is simple, robust , effective and easy to implement in rendering the Thangka line drawing.
出处 《科学技术与工程》 北大核心 2014年第16期107-111,118,共6页 Science Technology and Engineering
基金 国家自然基金项目(F010408) 西藏民族学院重大培育项目(13myZP04) 西藏自治区重大项目(12KJZRZMY01)资助
关键词 非真实感绘制 线条画 边缘切向流 基于流的高斯差分滤波 non-photorealistic rendering (NPR)line drawingedge tangential flow (ETF) flowbased difference of Gaussians (FDoG)
  • 相关文献

参考文献13

  • 1De Carlo D, Santella A. Stylization and abstraction of photographs.Proc ACM SIGGRAPH * 02, San Antonio Texas, USA, 2002:769-776.
  • 2Wang J, Xu Y, Shum, et al. Video tooning. Proc ACM * 04 TransGraphics,Los Angeles, 2004:23( 13) :574-583.
  • 3Collomosse J P,Rowntree D, Hall P M Stroke surfaces : temporallycoherent artistic animations &om video. IEEE Trar^ VisualizeUion andComputer Graphics,2005; 11 ( 5 ) : 540-549.
  • 4Wen F,Luan Q, Liang L, et al. Color sketch generation. Proc Non-Photorealistic Animation and Rendering ( NPAH * 06 ), Annecy ’France, 2006:47-54.
  • 5Fisher J, Bartz D, Strader W. Stylized au^nented reality for im-proved immersion. Proc IEEE Virtual Reality ( VR ’ 05〉,BonnGer-many,2005 : 195-202.
  • 6Tomi C,Manduchi R. Bilateral filtering for gray and color images.Proc IEEE Int’l Conf Computer Vision (ICCV ’98) IEEE CSPress,1998:839-?46.
  • 7Oraan A, Bousseau A, BarLa P, et al. Structure-preserving manipula-tion of photographs. Proc Non-Photorealistic Animation and Render-ing (NPAR ’07),ACM, New York ,2007 : 103-110.
  • 8Kai H, Chui C, Chakraborty U. A unified heme for adaptivestroke-ba8ed rendering. The Visual Computer, 2006 ; 22 ( 9 ):814-824.
  • 9Gooch B,Reinhard E,Gooch A. Human facial illustrations. ACMTrans Graphics,2004;23( 1) : 27-44.
  • 10Winnemo'* Her H, Ol^n S, Gooch B. Real-time video abstraction.Proc. ACM Transactions on Graphics ( SIGGRAPH 2006),BratonMassachusetts, USA ,2006: 1221_1226.

同被引文献16

引证文献2

二级引证文献2

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部