- 16
- 0
- 约5.98万字
- 约 21页
- 2019-01-05 发布于四川
- 举报
图形学专题
王山东等:图像抽象化的实时增强型绘制
PAGE 20
PAGE 19
图像抽象化的实时增强型绘制? Supported by a National Fundamental Research Grant of Science and Technology(973 Project:2009CB320802) and a research grant from the University of Macau.作者简介
? Supported by a National Fundamental Research Grant of Science and Technology(973 Project:2009CB320802) and a research grant from the University of Macau.作者简介:王山东(1985-),男,河南省夏邑县人,博士研究生,主要研究领域为计算机图形学,非真实绘制;李晓生(1986-),男,博士研究生,主要研究领域为计算机图形学;刘学慧(1968-),女,博士,副研究员,主要研究领域为计算机图形学;吴恩华(1947-),男,博士,研究员,博士生导师,主要研究领域为计算机图形学.
王山东1,2,3+, 李晓生1,3, 刘学慧1, 吴恩华1,2,3
1(中国科学院软件研究所计算机科学国家重点实验室 北京 100190)
2(澳门大学科技学院电脑与资讯科学系 澳门)
3(中国科学院研究生院 北京 100190)
摘 要:借鉴基于特征流的各向异性滤波框架,本文提出了一种增强对比度的图像抽象化绘制算法。针对输入图像,首先构造一个光滑连续的边缘切向流场,然后基于该流场对图像进行各向异性的增强处理,最后使用改进的Kuwahara滤波算法或其他特征保持的平滑滤波算法对图像进行实时处理。实验结果表明,本文算法可显著改善已有抽象化方法的绘制质量,抽象化图像的边界特征明显增强,画质更清晰,层次感更分明。
关键词:图像抽象化;边缘切向流;双边滤波;Kuwahara滤波
中图法分类号:TP391 文献标识码:A
Real-time Enhancing Image Abstraction
Wang Shandong1,2,3+, Li Xiaosheng1,3, Liu Xuehui1, Wu Enhua1,2,3
1(State Key Laboratory of Computer Science, Institute of Software Chinese Academy of Sciences, Beijing, 100190)
2(Department of Computer and Information Science, Faculty of Science and Technology, University of Macau, Macau)
3(Graduate University of Chinese Academy of Sciences, Beijing, 100190)
+ Corresponding author:Phone E-mail:sdwang10@
Abstract: Based on the flow-based anisotropic filtering framework, this paper presents a method for abstracting images with high color contrast. A smooth edge tangent flow field is firstly constructed, and then use the flow field to guide the process of enhancing local contrast, and at last an improved Kuwahara filter or other features-preserving filters are performed to achieve the abstraction result. Experimental results show that the proposed algorithm outperforms the existing methods in terms of feature preservation and visual appreciation.
Key words: image abstraction; edge tangent flow; bilateral filter; Kuwahara fil
原创力文档

文档评论(0)