一种基于SVD的图像水彩风格绘制算法

详细信息    查看全文 | 推荐本文 |

英文篇名：IMAGE WATERCOLOR STYLE-RENDERING ALGORITHM BASED ON SINGULAR VALUE DECOMPOSITION 作者：程琳琳 ; 陈昭炯 ; 傅明建 英文作者：Cheng Linlin;Chen Zhaojiong;Fu Mingjian;Guangdong Communications Polytechnic;College of Mathematics and Computer Science,Fuzhou University;National Experimental Teaching Demonstration Center of Network Information Security and Computer Technology,Fuzhou University; 关键词：奇异值分解(SVD) ; 色彩传递 ; mean-shift ; 水彩风格绘制 ; 分水岭变换 英文关键词：Singular value decomposition(SVD);;Color transfer;;Mean-shift;;Watercolor style-rendering;;Watershed conversion 中文刊名：JYRJ 英文刊名：Computer Applications and Software 机构：广东交通职业技术学院;福州大学数学与计算机科学学院;福州大学网络信息安全与计算机技术国家级实验教学示范中心; 出版日期：2019-05-12 出版单位：计算机应用与软件 年：2019 期：v.36 基金：国家自然科学基金项目(61772005);; 国家青年科学基金项目(61702104);; 广东省青年创新人才项目(2017GkQNCX022);; 广东省教育科研项目(2017GGXJK025);; 广东省攀登计划项目(pdjhb0750);; 教育部职业院校信息化教指委项目(2018LXB0302);; 福建省自然科学基金项目(2016J01295) 语种：中文; 页：JYRJ201905032 页数：5 CN：05 ISSN：31-1260/TP 分类号：189-192+221

摘要

针对传统图像水彩风格生成算法时间效率低,且容易出现色彩不协调的问题,提出一种基于奇异值分解SVD(singular value decomposition)的图像水彩风格绘制算法。利用SVD算法提取图像的主成分,再通过mean-shift算法实现图像的水彩风格效果。采用色彩传递算法,改善图像水彩风格的色彩视觉效果。实验结果表明,该算法实现的水彩风格艺术效果良好,可提高时间效率,同时色彩传递技术也使得水彩风格的艺术效果更加灵活多变。
        Aiming at the problem of low time efficiency and colour deviation in traditional watercolour style rendering algorithm, we proposed a watercolour style-rendering algorithm based on singular value decomposition(SVD). SVD algorithm was used to extract the principal components of image, and then mean-shift algorithm was used to achieve the watercolour style effect. We used colour transfer algorithm to improve the visual effect of watercolour style rendering. Experimental results show that the algorithm has good artistic effect and improves the time efficiency. Colour transfer technology also makes the artistic effect of watercolour style more flexible.

引文

[1] 陈为,张海嵩,于金辉.利用GPU实时绘制水墨画效果[J].计算机辅助设计与图形学学报,2005,17(11):2427-2432.
    [2] Fang W,Qing L,Lin L,et al.Color Sketch Generation[C]//Proceedings of the 4th International Symposium on Non-Photorealistic Animation and Rendering.Annecy,France:[s.n.],2006:47-54.
    [3] 梅洪,陈昭炯.基于Mean Shift和FDoG的图像卡通化渲染[J].计算机工程与应用,2016,52(10):213-217.
    [4] Curtis C J,Anderson S E,Seims J E,et al.Computer-generated watercolor[C]//Proceedings of the 24th annual conference on Computer graphics and interactive techniques.ACM,1997:421-429.
    [5] Johan H,Hashimoto R,Nishita T.Creating Watercolor Style Images Taking Into Account Painting Techniques[J].The Journal of the Society for Art and Science,2004,3(4):207-215.
    [6] 乔丽莎.基于深度学习的图像风格艺术化[D].西安:西安理工大学,2018.
    [7] Dan K.A singularly valuable decomposition:The SVD of a matrix[J].College Mathematics Journal,1996,27(1):2-23.
    [8] Reinhard E,Ashikhmin M,Gooch B,et al.Color Transfer Between Images[J].IEEE Computer Graphics and Applications,2001,21(5):34-41.
    [9] 杨静,陈昭炯.结合分水岭的Mean-Shift图像分割算法[J].小型微型计算机系统,2009,30(12):2493-2496.
    [10] Comaniciu D,Meer P.Mean-Shift:A Robust Approach Toward Feature Space Analysis[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2002,24(5):603-619.