Deep Learning Approaches to Classification of Production Technology for 19th Century Books

Cultural research is dedicated to understanding the processes of knowledge dissemination and the social and technological practices in the book industry. Research on children books in the 19th century can be supported by computer systems. Specifically, the advances in digital image processing seem to offer great opportunities for analyzing and quantifying the visual components in the books. The production technology for illustrations in books in the 19th century was characterized by a shift from wood or copper engraving to lithography. We report classification experiments which intend to classify images based on the production technology. For a classification task that is also difficult for humans, the classification quality reaches only around 70%. We analyze some further error sources and identify reasons for the low performance.

[1]  Iasonas Kokkinos,et al.  Deep Filter Banks for Texture Recognition, Description, and Segmentation , 2015, International Journal of Computer Vision.

[2]  Alexander Dunst,et al.  Auf dem Weg zur Visuellen Stilometrie:Automatische Genre- und Autorunterscheidung in graphischen Narrativen , 2018 .

[3]  Ali Farhadi,et al.  You Only Look Once: Unified, Real-Time Object Detection , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[4]  Paul F. Whelan,et al.  Using filter banks in Convolutional Neural Networks for texture classification , 2016, Pattern Recognit. Lett..

[5]  Shohreh Kasaei,et al.  Benign and malignant breast tumors classification based on region growing and CNN segmentation , 2015, Expert Syst. Appl..

[6]  Dumitru Erhan,et al.  Going deeper with convolutions , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Babak Saleh,et al.  A Unified Framework for Painting Classification , 2015, 2015 IEEE International Conference on Data Mining Workshop (ICDMW).

[8]  Pierre Bonnet,et al.  Plant Identification Based on Noisy Web Data: the Amazing Performance of Deep Learning (LifeCLEF 2017) , 2017, CLEF.

[9]  Rémy Mullot,et al.  A texture-based pixel labeling approach for historical books , 2015, Pattern Analysis and Applications.

[10]  Rémy Mullot,et al.  Texture feature benchmarking and evaluation for historical document image analysis , 2017, International Journal on Document Analysis and Recognition (IJDAR).