[1]王 征,李皓月,许洪山,等.基于卷积神经网络和SVM的中国画情感分类[J].南京师范大学学报(自然科学版),2017,40(03):74.[doi:10.3969/j.issn.1001-4616.2017.03.011]
 Wang Zheng,Li Haoyue,Xu Hongshan,et al.Chinese Painting Emotion Classification Based onConvolution Neural Network and SVM[J].Journal of Nanjing Normal University(Natural Science Edition),2017,40(03):74.[doi:10.3969/j.issn.1001-4616.2017.03.011]





Chinese Painting Emotion Classification Based onConvolution Neural Network and SVM
王 征1李皓月1许洪山1孙美君2
(1.大数据分析与系统实验室(天津大学软件学院),天津 300350)(2.天津大学计算机科学与技术学院,天津 300350)
Wang Zheng1Li Haoyue1Xu Hongshan1Sun Meijun2
(1.Laboratory for Big Data Analysis and System(School of Software,Tianjin University),Tianjin 300350,China)(2.School of Computer Science and Technology,Tianjin University,Tianjin 300350,China)
image emotionChinese paintingconvolution neural networkfeature extractionsupport vector machine
图像情感是指计算机识别数字图像所表达内容引起人的情感反应,根据不同的情感反应,可以对不同的图像进行分类. 在信息量急剧增长的今天,图像情感分类有助于图像的标注和检索,蕴藏着很大的社会和商业价值. 不同于西洋画的“以形写形”,中国画有着自己明显的特征:传统的国画不讲焦点透视,不强调自然界对于物体的光色变化,不拘泥于物体外表的肖似,而多强调抒发作者的主观情趣. 这比弥合一般的低层特征和人类情感高层语义之间的鸿沟的难度更大. 基于卷积神经网络因为其具有结构简单、适应性强、训练参数少、连接点多等特点,可以直接输入原始图像,能够避免对图像进行复杂的前期预处理. 相比传统图像特征提取方法,卷积神经网络具有明显的优势. 本文的目的是利用卷积神经网络发掘低层特征和情感语义之间的联系,提取国画图像特征,对得到的特征进行PCA降维、归一化等操作后,利用支持向量机(SVM)分类器进行情感分类.
Image emotions are human emotional responses caused by the contents of digital images. Computers are able to classify different images according to different human emotional responses. With the rapid growth of the amount of information,image emotion classification will contribute to the image annotation and search producing great social and commercial value. Chinese paintings have obvious characteristics:traditional Chinese paintings do not focus on the perspective,and do not emphasize the light color changes of objects in nature,and do not rigidly adhere to the appearance of objects. They more focus on the expression of authors’ subjective consciousness making it harder to bridge the semantic gap between general low-level features and human emotions. The structure of convolutional neural network(CNN)is simple,yet its adaptability is strong. CNN also has less training parameters and more junctions,and are able to read images directly without preprocessing images complexly. It has a huge advantage over traditional image-processing method. This paper aims to explore the relationships between low-level features and emotional semantics by CNN,and extract the features of Chinese paintings and process the features by PCA and normalization. Finally we classify the features by SVM.


[1] 柴刚. 解析国画作品中的意象世界[J]. 科技资讯,2009(20):197-198.
[2]雷莹. 浅析“情感”在国画创作中的艺术精神[J]. 大众文艺(学术版),2016(8):91.
[3]WRIGHT J,MA Y,MAIRAL J,et al. Sparse representation for computer vision and pattern recognition[J]. Proceedings of the IEEE,2010,98(6):1 031-1 044.
[4]SUN Y,FISHER R. Object-based visual attention for computer vision[J]. Artificial intelligence,2003,146(1):77-123.
[5]DATTA R. Semantics and aesthetics inference for image search:statistical leaning approaches[D]. University Park:The Pernnsylvania State University,2009.
[6]GUDIVADA V N. A geometry-based representation for efficient and effective retrieval of images by spatial similarity[J]. IEEE transactions of knowledge and data engineering,1998,10(3):504-512.
[7]HANJALIC,ALAN. Video and image retrieval beyond the cognitive level: the needs and possibilities[C]//Storage and Retrieval for Media Databases,San Jose,2001. Bellingham:Optical Engineering,2001:130-140.
[8]LECUN Y,BENGIO Y,HINTON G. Deep learning[J]. Nature,2015,521(7 553):436-444.
[9]HEMPHILL M. A note on adults’ color-emotion associations[J]. Journal of genetic psychology,1996,157(3):275-280.
[10]ELLIOT A J,MAIER M A,MOLLER A C,et al. Color and psychological functioning:the effect of red on performance attainment[J]. Journal of experimental psychology general,2007,136(1):154.
[11]陈俊杰. 图像情感语义分析技术[M]. 北京:电子工业出版社,2011:13-15.
[12]余英林,田菁,蔡志峰. 图像视觉感知信息的初步研究[J]. 电子学报,2001,29(10):1 373-1 375.
[13]WANG T,WU D J,COATES A,et al. End-to-end text recognition with convolutional neural networks[C]//International Conference on Pattern Recognition. Tsukuba Japan,2012. New York:IEEE,2012:3 304-3 308.
[14]ABDEL H O,MOHAMED A R,JIANG H,et al. Convolutional neural networks for speech recognition[J]. IEEE/ACM transactions on audio speech and language processing,2014,22(10):1 533-1 545.
[15]CECOTTI H,GR?SER A. Convolutional neural networks for P300 detection with application to brain-computer interfaces[J]. IEEE transactions on pattern analysis and machine intelligence,2011,33(3):433-45.
[16]NEBAUER C. Evaluation of convolutional neural networks for visual recognition[J]. IEEE transactions on neural networks,1998,9(4):685.
[17]BAI S. Growing random forest on deep convolutional neural networks for scene categorization[M]. London:Pergamon Press,2017:50-70.
[18]SUN M,SONG Z,JIANG X,et al. Learning pooling for convolutional neural network[J]. Neurocomputing,2017,224:96-104.
[19]YU S,JIA S,XU C. Convolutional neural networks for hyperspectral image classification[J]. Neurocomputing,2017,219:88-98.
[20]KUMAR A,KIM J,LYNDON D,et al. An ensemble of fine-tuned convolutional neural networks for medical image classification[J]. IEEE journal of biomedical and health informatics,2017,21(99):1.
[21]乐毅,王斌. 深度学习-Caffe之经典模型详解与实战[M]. 北京:电子工业出版社,2016:107-111.
[22]ZENG R,WU J,SHAO Z,et al. Quaternion softmax classifier[J]. Electronics letters,2014,50(25):1 929-1 931.
[23]CHEN X,YE Q,ZOU J,et al. Visual trajectory analysis via replicated softmax-based models[J]. Signal,image and video processing,2014,8(1):183-190.
[24]LEE J Y,KIM K. A feature-based approach to extracting machining features[J]. Computer-aided design,1998,30(13):1 019-1 035.
[25]CANDèS E J,LI X,MA Y,et al. Robust principal component analysis?[J]. Journal of the Acm,2009,58(3):11.
[26]XIE X,ZHENG W S,LAI J,et al. Normalization of face illumination based on large- and small-scale features[J]. IEEE transactions on image processing a publication of the IEEE signal processing society,2011,20(7):1 807-1 821.
[27]QI Z,TIAN Y,SHI Y. Robust twin support vector machine for pattern classification[J]. Pattern recognition,2013,46(1):305-316.
[28]QI Z,TIAN Y,SHI Y. Structural twin support vector machine for classification[J]. Knowledge-based systems,2013,43(2):74-81.
[29]SRIVASTAVA N,HINTON G,KRIZHEVSKY A,et al. Dropout:a simple way to prevent neural networks from overfitting[J]. Journal of machine learning research,2014,15(1):1 929-1 958.
[30]JR S E. Detecting and evaluating the impact of multidimensionality using item fit statistics and principal component analysis of residuals[J]. Journal of applied measurement,2002,3(2):205-231.


通讯联系人:孙美君,副教授,研究方向:图形学、图像及光谱数据处理. E-mail:sunmeijun@tju.edu.cn
更新日期/Last Update: 2017-09-30