Classification and indexing of paintings based on art movements

This paper outlines the automatic extraction of features of paintings´ art movements such as classicism, impressionism and cubism; and introduces a system developed for the classification and indexing of paintings based on their art movements. A six dimensional feature set is proposed for the representation of content and it is shown that the feature set enables to highlight art movements efficiently. In the classifier design, statistical pattern recognition approach is exploited and Bayesian, k-NN and SVM classifiers are employed. A classification accuracy of over 90% is achieved with very small false alarm ratios while the lowest performance is obtained by the k-NN. System also offers a quick query based database search by indexing the paintings with their six dimensional feature vectors, and provides an applicable program for museums and exhibition centres.