-
摘要:
图像艺术美感自动分类是近年的热门研究领域，国画作为中国传统艺术文化的重要体现，其美感也极具研究价值。在5类美感标注的国画数据库基础上，进行了国画艺术美感自动分类研究和相关特征分析。经过特征提取和筛选，得到适用于美感分类的33个图像特征，并基于特征重要性建立了物理特征与艺术美感、美术技法之间的映射关系。同时使用该特征集在多种分类器上进行艺术美感自动识别，验证了国画艺术美感自动分类的可行性。结果表明，国画艺术美感分类的主要相关美术元素按重要性排序为：颜色、笔触、亮度和线条。
Abstract:Automatic classification of aesthetics in images has been a popular research field in these years. Chinese traditional painting is a pivotal embodiment of Chinese traditional arts, so its aesthetics shows a great potential for researching. In this paper, the automatic classification study and relevant feature analysis of aesthetics were conducted in a Chinese painting database annotated with 5 classes of aesthetics. First, based on subjective annotation, by employing feature extraction and selection, 33 optimal image features were filtered out for aesthetic classification. Then, a mapping analysis was conducted on the relationship among objective features, subjective aesthetics and image artistic elements. Finally, an automatic recognition using a variety of mainstream classifiers was implemented on the optimal feature set, and an acceptable performance was obtained, which proves the feasibility and effectiveness of automatic classification of Chinese painting aesthetics. The results show that the main artistic elements (in order) of aesthetic classification for Chinese traditional painting are:color, brushwork, brightness and lines.
-
图 7 国画美感分类的美术元素重要性
Figure 7. Importance of artistic element in aesthetic classification of Chinese traditional painting
表 1 图画美感分类数量
Table 1. Number of paintings in each aesthetic classification
美感 山水画 花鸟画 总数目 气势美 110 1 111 清幽美 44 9 53 生机美 54 133 187 雅致美 11 92 103 萧瑟美 30 13 43 无法分类 6 8 14 总数目 255 256 511 
下载: 导出CSV
表 2 有效特征及原理
Table 2. Effective features and theories
特征 具体维度 美术或图像原理 色相直方图 H4, H9, H14 颜色 颜色简明度 Hs 颜色 红色直方图 R7, R10 颜色 绿色直方图 G6 颜色 蓝色直方图 B10 颜色 灰度直方图 Gr1, Gr4, Gr5, Gr12, Gr16 亮度、颜色 对比度 C 亮度和颜色 亮度直方图 V2, V3, V5 亮度 暗通道 D2 亮度 图像模糊度 B 模糊度、笔触 邻域相似性 N3, N4, N7, N10, N11, N12, N16, N17, N18, N24 笔触 直线段个数 L 线条、笔触 边界复杂度 E9, E10 线条、构图 显著性图样 S 视觉注意、亮度
下载: 导出CSV
表 3 各美感有效特征及原理
Table 3. Effective features and theories for each item of aesthetics
特征 气势美 清幽美 生机美 雅致美 萧瑟美 色相直方图 √ √ √ √ 颜色简明度 √ √ 红色直方图 √ √ √ 绿色直方图 √ √ √ 蓝色直方图 √ √ √ 灰度直方图 √ √ √ 对比度 √ √ 亮度直方图 √ √ √ 暗通道 √ √ √ 图像模糊度 √ 邻域相似性 √ √ √ √ 直线段个数 √ √ √ 边界复杂度 √ √ 显著性图样 √
下载: 导出CSV
表 4 不同分类器下的美感自动分类结果
Table 4. Automatic classification results of aesthetics in different classifiers
美感 Extra-Trees SVM 线性判别分析 随机森林 KNN 朴素贝叶斯 逻辑回归 多元感知机 P R P R P R P R P R P R P R P R 气势美 0.57 0.63 0.54 0.66 0.57 0.64 0.49 0.61 0.47 0.60 0.34 0.53 0.56 0.65 0.54 0.62 生机美 0.53 0.74 0.58 0.64 0.57 0.67 0.50 0.66 0.55 0.66 0.58 0.18 0.57 0.71 0.57 0.64 雅致美 0.52 0.41 0.51 0.44 0.55 0.48 0.48 0.38 0.51 0.44 0.46 0.44 0.49 0.45 0.44 0.45 萧瑟美 0.20 0.04 0.43 0.23 0.41 0.14 0.20 0.07 0 0 0.18 0.47 0.42 0.14 0.39 0.14 清幽美 0.49 0.21 0.34 0.21 0.34 0.25 0.51 0.21 0.38 0.29 0.16 0.09 0.34 0.16 0.36 0.27 平均 0.46 0.41 0.48 0.44 0.49 0.44 0.44 0.39 0.38 0.40 0.34 0.34 0.48 0.42 0.46 0.42
下载: 导出CSV
表 6 国画美感特征重要性系数
Table 6. Importance coefficient of aesthetic features in Chinese traditional painting
特征 重要性系数 颜色简明度 0.1756 边界复杂度10 0.0945 直线段个数 0.0896 亮度直方图 5 0.0548 红色直方图 10 0.0473 红色直方图 7 0.0402 亮度直方图 3 0.0357 灰度直方图 12 0.0264 绿色直方图 6 0.0259 邻域相似性18 0.0244 灰度直方图 5 0.0231 邻域相似性24 0.0213 亮度直方图 2 0.0207 邻域相似性3 0.0205 邻域相似性11 0.0200 邻域相似性7 0.0200 边界复杂度9 0.0195 暗通道2 0.0194 邻域相似性12 0.0191 灰度直方图 1 0.0191 灰度直方图 16 0.0186 图像模糊度 0.0174 显著性区域均值 0.0169 灰度直方图 4 0.0164 颜色直方图 4 0.0159 邻域相似性4 0.0154 蓝色直方图 10 0.0150 对比度 0.0148 邻域相似性17 0.0140 邻域相似性16 0.0129 邻域相似性10 0.0102 颜色直方图 14 0.0092 颜色直方图 9 0.0064
下载: 导出CSV
-
[1] JOSHI D, DATTA R.Aesthetics and emotions in images[J]. IEEE Signal Processing Magazine, 2011, 28(5):94-115. doi: 10.1109/MSP.2011.941851 [2] PERRONNIN F, MARCHESOTTI L, MURRAY N.AVA: A large-scale database for aesthetic visual analysis[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Piscataway, NJ: IEEE Press, 2012: 2408-2415. [3] LUO W, WANG X, TANG X.Content-based photo quality assessment[C]//IEEE International Conference on Computer Vision.Piscataway, NJ: IEEE Press, 2012: 2206-2213. [4] LI C, CHEN T.Aesthetic visual quality assessment of paintings[J]. IEEE Journal of Selected Topics in Signal Processing, 2009, 3(2):236-252. doi: 10.1109/JSTSP.2009.2015077 [5] MOHAMMAD S M, TURNEY P D.WikiArt emotions: An annotated dataset of emotions evoked by art[C]//Proceedings of the 11th Edition of the Language Resources and Evaluation Conference, 2018. [6] MENSINK T, VAN GEMERT J C.The Rijksmuseum challenge: Museum-centered visual recognition[C]//Proceedings of International Conference on Multimedia Retrieval.New York: ACM, 2014: 451-454. [7] KE Y, TANG X, JING F, et al.The design of high-level features for photo quality assessment[C]//Proceedings of IEEE Comference on Computer Vision and Pattern Recognition.Piscataway, NJ: IEEE Press, 2006: 419-426. [8] LUO Y, TANG X.Photo and video quality evaluation: Focusing on the subject[C]//Proceedings of European Conference on Computer Vision.Berlin: Springer, 2008: 386-399. [9] WU Y, BAUCKHAGE C, THURAU C, et al.The good, the bad, and the ugly: Predicting aesthetic image labels[C]//Proceedings of International Conference on Pattern Recognition.Piscataway, NJ: IEEE Press, 2010: 1586-1589. [10] 陈俊杰, 杜雅娟, 李海芳.中国画的特征提取及分类[J].计算机工程与应用, 2008, 44(15):166-169. doi: 10.3778/j.issn.1002-8331.2008.15.052CHEN J J, DU Y J, LI H F.Feature extraction and classification of Chinese painting[J]. Computer Engineering and Applications, 2008, 44(15):166-169(in Chinese). doi: 10.3778/j.issn.1002-8331.2008.15.052 [11] 刘晓巍, 普园媛, 黄亚群, 等.绘画视觉艺术风格的量化统计与分析[J].计算机科学与探索, 2013, 7(10):962-972. http://d.old.wanfangdata.com.cn/Periodical/jsjkxyts201310008LIU X W, PU Y Y, HUANG Y Q, et al.Quantitative statistics and analysis for painting visual art style[J]. Journal of Frontiers of Computer Science and Technology, 2013, 7(10):962-972(in Chinese). http://d.old.wanfangdata.com.cn/Periodical/jsjkxyts201310008 [12] 王征, 孙美君, 韩亚洪, 等.监督式异构稀疏特征选择的国画分类和预测[J].计算机辅助设计与图形学学报, 2013, 25(12):1848-1855. http://d.old.wanfangdata.com.cn/Periodical/jsjfzsjytxxxb201312010WANG Z, SUN M J, HAN Y H, et al.Supervised heterogeneous sparse feature selection for Chinese paintings classification[J]. Journal of Computer-Aided Design & Computer Graphics, 2013, 25(12):1848-1855(in Chinese). http://d.old.wanfangdata.com.cn/Periodical/jsjfzsjytxxxb201312010 [13] 盛家川, 李玉芝.国画的艺术目标分割及深度学习与分类[J].中国图象图形学报, 2018, 23(8):1193-1206. http://d.old.wanfangdata.com.cn/Periodical/zgtxtxxb-a201808009SHENG J C, LI Y Z.Learning artistic objects for improved classification of Chinese paintings[J]. Journal of Image and Graphics, 2018, 23(8):1193-1206(in Chinese). http://d.old.wanfangdata.com.cn/Periodical/zgtxtxxb-a201808009 [14] 高峰, 聂婕, 黄磊, 等.基于表现手法的国画分类方法研究[J].计算机学报, 2017, 40(12):2871-2882. doi: 10.11897/SP.J.1016.2017.02871GAO F, NIE J, HUANG L, et al.Traditional Chinese painting classification based on painting technique[J]. Chinese Journal of Computers, 2017, 40(12):2871-2882(in Chinese). doi: 10.11897/SP.J.1016.2017.02871 [15] 李玉芝, 盛家川, 华斌.中国画分类的改进嵌入式学习算法[J].计算机辅助设计与图形学学报, 2018, 30(5):893-900. http://d.old.wanfangdata.com.cn/Periodical/jsjfzsjytxxxb201805017LI Y Z, SHENG J C, HUA B.Improved embedded learning for classification of Chinese paintings[J]. Journal of Computer-Aided Design & Computer Graphics, 2018, 30(5):893-900(in Chinese). http://d.old.wanfangdata.com.cn/Periodical/jsjfzsjytxxxb201805017 [16] 张佳婧, 彭韧, 王健, 等.水墨画计算审美评估[J].软件学报, 2016, 27(增刊2):220-233.ZHANG J J, PENG R, WANG J, et al.Computational aesthetic evaluation of Chinese wash paintings[J]. Journal of Software, 2016, 27(Suppl.2):220-233(in Chinese). [17] 陈丽君.美感与积极情绪的关系及对变化觉察的影响[D].重庆: 西南大学, 2010.CHEN L J.The relationship between aesthetic experience and positive emotion and the impact on change detection[D]. Chongqing: Southwest University, 2010(in Chinese). [18] ISRAELI N.Affective reactions to painting reproductions:A study in the psychology of esthetics[J]. Journal of Applied Psychology, 1928, 12(1):125-139. [19] HAGTVEDT H, HAGTVEDT R, PATRICK V M.The perception and evaluation of visual art[J]. Empirical Studies of the Arts, 2008, 26(2):197-218. doi: 10.2190/EM.26.2.d [20] STAMATOPOULOU D.Integrating the philosophy and psychology of aesthetic experience:Development of the aesthetic experience scale[J]. Psychological Reports, 2004, 95(2):673-695. [21] SILVIA P J, FAYN K, NUSBAUM E C, et al.Openness to experience and awe in response to nature and music:Personality and profound aesthetic experiences[J]. Psychology of Aesthetics Creativity & the Arts, 2015, 9(4):376-384. [22] MARKOVIC' S.Aesthetic experience and the emotional content of paintings[J]. Psihologija, 2010, 43(1):47-64. doi: 10.2298/PSI1001047M [23] ROWOLD J.Instrument development for esthetic perception assessment[J]. Journal of Media Psychology Theories Methods & Applications, 2008, 20(1):35-40. [24] HAGER M, HAGEMANN D, DANNER D, et al.Assessing aesthetic appreciation of visual artworks-The construction of the art reception survey (ARS)[J]. Psychology of Aesthetics Creativity & the Arts, 2012, 9(4):320-333. [25] KARINA V.Die emotionale Wirkung moderner Kunst[D]. Deutschland: Universität Wien, 2010. [26] 丁月华.概念隐喻理解中的美感体验对科学概念理解的作用研究[D].重庆: 西南大学, 2008.DING Y H.A research on the role of aesthetic experience of concept metaphor understanding to the scientific concept understanding[D]. Chongqing: Southwest University, 2008(in Chinese). [27] HEVNER K.Experimental studies of the elements of expression in music[J]. American Journal of Psychology, 1936, 48(2):246-268. doi: 10.2307-1415746/ [28] 孟子厚.音质主观评价的实验心理学方法[M].北京:国防工业出版社, 2008:84-89.MENG Z H.Experimental psychological method of subjective evaluation of sound quality[M]. Beijing:National Defense Industry Press, 2008:84-89(in Chinese). [29] CATTELL R B.The scientific use of factor analysis in behavioral and life sciences[M]. Berlin:Springer, 1978. [30] 湛颖, 高妍, 谢凌云.中国国画情感-美感数据库[J/OL].中国图像图形学报(2019-06-19)[2019-07-03].http://www.cjig.cn/jig/ch/reader/view_abstract.aspx?flag=2&file_no=201903210000001&journal_id=jig.ZHAN Y, GAO Y, XIE L Y.A database for emotion and aesthetic analysis on Chinese traditional paintings[J/OL]. Journal of Image and Graphics(2019-06-19)[2019-07-03]. http://www.cjig.cn/jig/ch/reader/view_abstract.aspx?flag=2&file_no=201903210000001&journal_id=jig(in Chinese). [31] HE K, SUN J, TANG X, et al.Single image haze removal using dark channel prior[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition.Piscataway, NJ: IEEE Press, 2009: 1956-1963. [32] CANNY J F.A computational approach to edge detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1986, 8(6):679-698. http://d.old.wanfangdata.com.cn/OAPaper/oai_doaj-articles_3d39d0b11988c5f90bf44b10d764f020 [33] DONG Z, TIAN X.Multi-level photo quality assessment with multi-view features[J]. Neurocomputing, 2015, 168(30):308-319. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=83cda8c0f4734e5c195f6102bb379c64 [34] KOCH C, ULLMAN S.Shifts in selective visual attention:Towards the underlying neural circuitry[J]. Human Neurobiology, 1987, 4(2):115-141. [35] GUYON I, WESTON J, BARNHILL S, et al.Gene selection for cancer classification using support vector machines[J]. Machine Learning, 2002, 46(1):389-422. http://d.old.wanfangdata.com.cn/OAPaper/oai_pubmedcentral.nih.gov_2216417 [36] 周志华.机器学习[M].北京:清华大学出版社, 2016:26-30.ZHOU Z H.Machine learning[M]. Beijing:Tsinghua University Press, 2016:26-30(in Chinese). [37] GEURTS P, ERNST D, WEHENKEL L, et al.Extremely randomized trees[J]. Machine Learning, 2006, 63(1):3-42. http://d.old.wanfangdata.com.cn/NSTLQK/NSTL_QKJJ0227516002/ -
下载:
点击查看大图
计量
- 文章访问数: 1088
- HTML全文浏览量: 200
- PDF下载量: 407
- 被引次数: 0
