Document Image Classification with Intra-Domain Transfer Learning and Stacked Generalization of Deep Convolutional Neural Networks
Document Type
Conference Article
Publication Title
Proceedings - International Conference on Pattern Recognition
Abstract
In this article, a region-based Deep Convolutional Neural Network framework is presented for document structure learning. The contribution of this work involves efficient training of region based classifiers and effective ensembling for document image classification. A primary level of 'inter-domain' transfer learning is used by exporting weights from a pre-trained VGG16 architecture on the ImageNet dataset to train a document classifier on whole document images. Exploiting the nature of region based influence modelling, a secondary level of 'intra-domain' transfer learning is used for rapid training of deep learning models for image segments. Finally, a stacked generalization based ensembling is utilized for combining the predictions of the base deep neural network models. The proposed method achieves state-of-the-art accuracy of 92.21% on the popular RVL-CDIP document image dataset, exceeding the benchmarks set by the existing algorithms.
First Page
3180
Last Page
3185
DOI
10.1109/ICPR.2018.8545630
Publication Date
11-26-2018
Recommended Citation
Das, Arindam; Roy, Saikat; Bhattacharya, Ujjwal; and Parui, Swapan K., "Document Image Classification with Intra-Domain Transfer Learning and Stacked Generalization of Deep Convolutional Neural Networks" (2018). Conference Articles. 38.
https://digitalcommons.isical.ac.in/conf-articles/38
Comments
Open Access, Green