Image analysis for artistic style identification: A powerful tool for preserving cultural heritage

In this paper we propose a new adaptable neural network architecture able to identify the artistic style of paintings. This technique is a non-destruction method for preserving cultural heritage. Preserving visual paintings and restoring damages is one of the most salient actions in conserving work arts and antiquities. More specifically, the preservation procedure often follows not only current knowledge of the artist's common practices but they also include a visual assessment of the presence of the artist's "handwriting" in the brushwork. The features used are of two types; MPEG-7 visual descriptors and SIFT descriptors. We also propose scalable organization of the descriptors.