计算机科学与探索 ›› 2016, Vol. 10 ›› Issue (7): 1021-1034.DOI: 10.3778/j.issn.1673-9418.1510048

• 人工智能与模式识别 • 上一篇    下一篇

沙画图像特征的可计算美学分析

吴  涛1+,杨俊杰2,冉国洪3   

  1. 1. 岭南师范学院 信息科学与技术学院,广东 湛江 524048
    2. 广东高校数字化学习工程技术开发中心,广东 湛江 524048
    3. 岭南师范学院 美术学系,广东 湛江 524048
  • 出版日期:2016-07-01 发布日期:2016-07-01

Computational Aesthetics Analysis on Sand Painting Style

WU Tao1+, YANG Junjie2, RAN Guohong3   

  1. 1. School of Information Science and Technology, Lingnan Normal University, Zhanjiang, Guangdong 524048, China
    2. Guangdong Engineering and Technological Development Center for E-learning, Zhanjiang, Guangdong 524048, China
    3. Department of Fine Arts, Lingnan Normal University, Zhanjiang, Guangdong 524048, China
  • Online:2016-07-01 Published:2016-07-01

摘要: 沙画是一种风靡全球的新视觉艺术形式,计算机沙画艺术方兴未艾。在可计算美学的研究框架下,引入了色彩模式、主题色、色相环、Benford度量、分形维、全局对比因子、Shannon熵、彩色度、数据场复杂度、Tamura纹理等量化特征维,对沙画艺术的创作风格进行了数值计算和统计,探索了静态沙画图像的可计算美学分析。实验结果表明,计算机辅助的客观美学评价能够一定程度上反映不同沙画作品创作风格上的异同点,基本符合人类主观审美认知,作为构建计算机沙画艺术与沙画表演作品之间的有益桥梁,可为沙画艺术创作风格的主观评价提供客观依据和量化补充,也为计算机沙画艺术自动创作提供参考和借鉴。

关键词: 艺术风格, 沙画艺术, 可计算美学, 图像统计, 图像可视化

Abstract: Sand painting is a very new and popular performance art technique, and the researches on computer-aided sand painting are growing rapidly, but most of the current works seem to be biased. Various features, including color pattern, color theme, color wheel, Benford index, fractal dimension, global contrast factor, Shannon entropy, colorfulness, data field complexity and Tamura texture, are introduced into the framework of computational aesthetics, and several sand performers are involved into the numerical calculation and statistics of sand painting art style. Then the computational aesthetics analysis on sand images is achieved. The experimental results suggest that, the computer-aided aesthetic evaluations can reflect the differences and similarities among the different types of sand images, which well correspond to human aesthetic perception, and as one of the useful bridge between sand painting art and computer-aided sand animation, the quantitative supplement is proposed for the subjective evaluation of sand painting art style, and then a reference is also provided for automatic creation of computer-aided sand animation.

Key words: art style, sand painting art, computational aesthetics, image statistics, image visualization