Deepdocclassifier: Document classification with deep Convolutional Neural Network

• Published in: 2015 13th International Conference on Document Analysis and Recognition (ICDAR) pp. 1111 - 1115
• Main Authors: Afzal, Muhammad Zeshan, Capobianco, Samuele, Malik, Muhammad Imran, Marinai, Simone, Breuel, Thomas M., Dengel, Andreas, Liwicki, Marcus
• Format: Conference Proceeding
• Language: English
• Published: IEEE 01.08.2015
• Subjects: Convolutional codes; Convolutional Neural Network; Deep CNN; Document Image Classification; Electronic mail; Marine vehicles
• DOI: 10.1109/ICDAR.2015.7333933

This paper presents a deep Convolutional Neural Network (CNN) based approach for document image classification. One of the main requirement of deep CNN architecture is that they need huge number of samples for training. To overcome this problem we adopt a deep CNN which is trained using big image dataset containing millions of samples i.e., ImageNet. The proposed work outperforms both the traditional structure similarity methods and the CNN based approaches proposed earlier. The accuracy of the proposed approach with merely 20 images per class outperforms the state-of-the-art by achieving classification accuracy of 68.25%. The best results on Tobbacoo-3428 dataset show that our proposed method outperforms the state-of-the-art method by a significant margin and achieved a median accuracy of 77.6% with 100 samples per class used for training and validation.