A unified framework for document restoration using inpainting and shape-from-shading

  • Authors:

  • Li Zhang;A. M. Yip;M. S. Brown;Chew Lim Tan

  • Affiliations:

  • School of Computing, National University of Singapore, Singapore;Department of Mathematics, National University of Singapore, Singapore;School of Computing, National University of Singapore, Singapore;School of Computing, National University of Singapore, Singapore

  • Venue:
  • Pattern Recognition
  • Year:
  • 2009

Quantified Score

Hi-index 0.01

Visualization

Abstract

We present a restoration framework to reduce undesirable distortions in imaged documents. Our framework is based on two components: (1) an image inpainting procedure that can separate non-uniform illumination (and other) artifacts from the printed content and (2) a shape-from-shading (SfS) formulation that can reconstruct the 3D shape of the document's surface. Used either piecewise or in its entirety, this framework can correct a variety of distortions including shading, shadow, ink-bleed, show-through, perspective and geometric distortions, for both camera-imaged and flatbed-imaged documents. Our overall framework is described in detail. In addition, our SfS formulation can be easily modified to target various illumination conditions to suit different real-world applications. Results on images of synthetic and real documents demonstrate the effectiveness of our approach. OCR results are also used to gauge the performance of our approach.