End-to-End Unsupervised Document Image Blind Denoising
This addresses the challenge of document denoising for OCR systems in real-world settings where supervised data is scarce, though it appears incremental as it builds on existing unsupervised deep learning approaches.
The paper tackles the problem of removing multiple types of noise from scanned document images without requiring paired noisy/clean data, and demonstrates that the proposed model significantly improves image quality and OCR accuracy on several test datasets.
Removing noise from scanned pages is a vital step before their submission to the optical character recognition (OCR) system. Most available image denoising methods are supervised where the pairs of noisy/clean pages are required. However, this assumption is rarely met in real settings. Besides, there is no single model that can remove various noise types from documents. Here, we propose a unified end-to-end unsupervised deep learning model, for the first time, that can effectively remove multiple types of noise, including salt \& pepper noise, blurred and/or faded text, as well as watermarks from documents at various levels of intensity. We demonstrate that the proposed model significantly improves the quality of scanned images and the OCR of the pages on several test datasets.