A Novel Method for Emotion Extraction From Paintings Based on Luscher’s Psychological Color Test: Case Study Iranian-Islamic Paintings

Paintings evoke certain emotions in the viewers. Colors, shape, texture, and many other factors affect the feeling conveyed by paintings, but colors seem to have a stronger effect due to a century-long study of color-emotion association in various fields of psychology, art, and color science. There are many color-emotion theories and most of them have been implemented, however, the Luscher Color Test is untouched amongst them. Based on several reasons, discussed in detail inside the paper, we believe this theory can cover problems in this domain of emotion extraction. The main motivation for choosing the Luscher test was that this method is designed for personality and mood analysis and it can better study abstract paintings. In this paper, a set of paintings from Iranian-Islamic cultural heritage is chosen as a dataset. We have proposed the L-EEP method based on Culture Technology (CT) concept to extract emotions from paintings with image processing techniques and psychology knowledge. This method extracts specific colors from paintings and by performing the Luscher test automatically, is able to determine eight emotions. For this matter, paintings are assessed in two moods: 1. The full extent of the painting 2. Cropped interest area of the painting that attracts more attention. Then, the color palette which is extracted colors ordered based on their coverage extent enters search engine. The search engine performs the searching process in the 3D knowledge base of Luscher color-emotion layers to extract relative values of emotions in both scenarios. For the evaluation of the results, three steps were taken. First, we compared the output results of ancient Persian painting with literature and text of their background stories. Then a viewer evaluation is done to compare the results with human viewpoint. Finally, a set of modern abstract paintings peer-rated in the IAPS standard system to further examine the proposed method. The results of the three forms of evaluation indicate the applicability of the L-EEP approach.

[1]  Alberto Del Bimbo,et al.  Semantics in Visual Information Retrieval , 1999, IEEE Multim..

[2]  Bryan Pardo,et al.  Classifying paintings by artistic genre: An analysis of features & classifiers , 2009, 2009 IEEE International Workshop on Multimedia Signal Processing.

[3]  Nader Jafarnia Dabanloo,et al.  Personality psychology using heart responses to color stimulus , 2011, 2011 Computing in Cardiology.

[4]  Farid Amir,et al.  THE STUDY AND ANALYSIS OF ICONOGRAGHY OF AN IMAGE THE DEATH OF SHIDAH BY KAY KHOSROW , 2013 .

[5]  P. Valdez,et al.  Effects of color on emotions. , 1994, Journal of experimental psychology. General.

[6]  Allan Hanbury,et al.  Affective image classification using features inspired by psychology and art theory , 2010, ACM Multimedia.

[7]  Takayuki Uchida,et al.  Color-Tunable Single Pixels Using Stacked Transparent Organic Light Emitting Diodes and Color-Tunable Lighting Domes , 2017 .

[8]  Markus A. Maier,et al.  Color psychology: effects of perceiving color on psychological functioning in humans. , 2014, Annual review of psychology.

[9]  L. Ou,et al.  A study of colour emotion and colour preference. Part I: Colour emotions for single colours , 2004 .

[10]  Kyunghyun Yoon,et al.  A method for extracting emotion using colors comprise the painting image , 2018, Multimedia Tools and Applications.

[11]  Dae-Won Kim,et al.  System for matching paintings with music based on emotions , 2016, SIGGRAPH Asia Technical Briefs.

[12]  Nicu Sebe,et al.  In the eye of the beholder: employing statistical analysis and eye tracking for analyzing abstract paintings , 2012, ACM Multimedia.

[13]  Nicu Sebe,et al.  Emotional valence categorization using holistic image features , 2008, 2008 15th IEEE International Conference on Image Processing.

[14]  L. Rainwater,et al.  The meanings of color. , 1962, The Journal of general psychology.

[15]  Wei-Ning Wang,et al.  Image emotional semantic query based on color semantic description , 2005, 2005 International Conference on Machine Learning and Cybernetics.

[16]  M. Lüscher,et al.  The Lüscher Color Test , 1969 .

[17]  Norouzi Nastaran,et al.  A SURVEY OF KAMAL-AL-DIN BEHZAD’S INTERPRETATION OF THE STORY OF “SA’DI AND JAVAN-E-KASHGHARI»; BASED ON THE MYSTICAL TRAINING PATTERN OF NUR-AL-DIN ABD-AL RAHMAN , 2015 .

[18]  Alan Wee-Chung Liew,et al.  Colour-emotion association study on abstract art painting , 2015, 2015 International Conference on Machine Learning and Cybernetics (ICMLC).

[19]  C. A. Joseph,et al.  Relationship between the Luscher Color Test and the MMPI. , 1984, Journal of clinical psychology.

[20]  Charles E. Osgood,et al.  A Cross-Cultural Study of the Affective Meanings of Color , 1973 .

[21]  Kyunghyun Yoon,et al.  Mood from painting: Estimating the mood of painting by using color image scale , 2015, 2015 21st Korea-Japan Joint Workshop on Frontiers of Computer Vision (FCV).

[22]  Lois B. Wexner.  The degree to which colors (hues) are associated with mood-tones. , 1954 .

[23]  A. B. Alexander,et al.  THE LUSCHER COLOR TEST: AN INVESTIGATION OF VALIDITY AND UNDERLYING ASSUMPTIONS , 1972 .

[24]  James Ze Wang,et al.  Image retrieval: Ideas, influences, and trends of the new age , 2008, CSUR.

[25]  J. Russell.  A circumplex model of affect. , 1980 .

[26]  L. Ou,et al.  A study of colour emotion and colour preference. Part II: Colour emotions for two‐colour combinations , 2004 .

[27]  Shigenobu Kobayashi,et al.  The aim and method of the color image scale , 2009 .

[28]  Naz Kaya,et al.  Relationship between color and emotion: A study of college students. , 2004 .

[29]  V. Solouk,et al.  Friend recommendation based on the Luscher color theory: Twitter use case , 2013, 2013 IEEE 11th Malaysia International Conference on Communications (MICC).

[30]  James Zijun Wang,et al.  Identifying Emotions Aroused from Paintings , 2016, ECCV Workshops.