Document Image Binarization with Fully Convolutional Neural Networks
This addresses the pre-processing need for document processing tasks in historical manuscript analysis, representing a strong specific gain in this domain.
The paper tackles the problem of binarizing degraded historical manuscript images by formulating it as a pixel classification task using a novel Fully Convolutional Network (FCN) architecture, achieving results that outperform competition winners on 4 out of 7 DIBCO competitions.
Binarization of degraded historical manuscript images is an important pre-processing step for many document processing tasks. We formulate binarization as a pixel classification learning task and apply a novel Fully Convolutional Network (FCN) architecture that operates at multiple image scales, including full resolution. The FCN is trained to optimize a continuous version of the Pseudo F-measure metric and an ensemble of FCNs outperform the competition winners on 4 of 7 DIBCO competitions. This same binarization technique can also be applied to different domains such as Palm Leaf Manuscripts with good performance. We analyze the performance of the proposed model w.r.t. the architectural hyperparameters, size and diversity of training data, and the input features chosen.