Folk art classification using support vector machine

Abstract

Tremendous  amounts  of  effort  have  been  carried  out  every  year  by  the governments of all the countries to preserve art and culture. Art in the form of  paintings,  artifacts,  music,  dance,  and  cuisines  of  every  country  has  the utmost  importance.  The  study  of Tribal  arts  provides  deep  insight  into  our history  and  acts  as  a  milestone  in  the  roadmap  of  our  future.  This  paper focuses on three popular folk arts namely: Gond, Manjusha, and Warli. 300 images of each artwork have been collected from various online repositories. To  generate  a  robust  system,  data  augmentation  is  appliedwhich  results  in 7510 images. A feature vector based on a generalized co-occurrence matrix, local binary pattern, HSV histogram, and canny edge detector is constructed and  classification  is  performed  using  a  linear  support  vector  machine.  10-fold cross-validation produces 99.8% accuracy.