Document rectification and illumination correction using a patch-based CNN
Authors:
Open Access
- 8 November 2019
- journal article
- research article
- Published by Association for Computing Machinery (ACM) in ACM Transactions on Graphics
- Vol. 38 (6) , 1-11
Abstract
We propose a novel learning method to rectify document images with various distortion types from a single input image. As opposed to previous learning-based methods, our approach seeks to first learn the distortion flow on input image patches rather than the entire image. We then present a robust technique to stitch the patch results into the rectified document by processing in the gradient domain. Furthermore, we propose a second network to correct the uneven illumination, further improving the readability and OCR accuracy. Due to the less complex distortion present on the smaller image patches, our patch-based approach followed by stitching and illumination correction can significantly improve the overall accuracy in both the synthetic and real datasets.Keywords
All Related Versions
This publication has 34 references indexed in Scilit:
- Transform recipes for efficient cloud photo enhancementACM Transactions on Graphics, 2015
- Document dewarping via text-line based optimizationPattern Recognition, 2015
- Image-based bidirectional scene reprojectionACM Transactions on Graphics, 2011
- State Estimation in a Document Image and Its Application in Text Block Identification and Text Line Extraction2010
- A unified framework for document restoration using inpainting and shape-from-shadingPattern Recognition, 2009
- Unwarping scanned image of Japanese/English documents2007
- Multi-View Document Rectification using Boundary2007
- Document Image De-warping Based on Detection of Distorted Text Lines2005
- Document capture using stereo vision2004
- Document Image De-warping for Text/Graphics Recognition2002