PDNet: Semantic Segmentation integrated with a Primal-Dual Network for Document binarization

Abstract

Binarization of digital documents is the task of classifying each pixel in an image of the document as belonging to the background (parchment/paper) or foreground (text/ink). Historical documents are often subjected to degradations, that make the task challenging. In the current work a deep neural network architecture is proposed that combines a fully convolutional network with an unrolled primal-dual network that can be trained end-to-end to achieve state of the art binarization on four out of seven datasets. Document binarization is formulated as an energy minimization problem. A fully convolutional neural network is trained for semantic segmentation of pixels that provides labeling cost associated with each pixel. This cost estimate is refined along the edges to compensate for any over or under estimation of the foreground class using a primal-dual approach. We provide necessary overview on proximal operator that facilitates theoretical underpinning required to train a primal-dual network using a gradient descent algorithm. Numerical instabilities encountered due to the recurrent nature of primal-dual approach are handled. We provide experimental results on document binarization competition dataset along with network changes and hyperparameter tuning required for stability and performance of the network. The network when pre-trained on synthetic dataset performs better as per the competition metrics.

arXiv commentUnder consideration for Pattern Recognition Letters Special Issue on Graphonomics for e-citizens: e-health, e-society, e-education 11 pages, 10 figures, 2 tables

References in corpus (11)

• Adam: A Method for Stochastic Optimization
• Deep Residual Learning for Image Recognition
• Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs
• Conditional Random Fields as Recurrent Neural Networks
• Proceedings of the 29th International Conference on Machine Learning (ICML-12)
• A Review on Deep Learning Techniques Applied to Semantic Segmentation
• ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation
• Rethinking the Inception Architecture for Computer Vision
• Flattened Convolutional Neural Networks for Feedforward Acceleration
• A Deep Primal-Dual Network for Guided Depth Super-Resolution
• Learning Document Image Binarization from Data

Cited by in corpus (2)

• Unsupervised Neural Domain Adaptation for Document Image Binarization
• Analysis of PDE-based binarization model for degraded document images