Image restoration of arbitrarily warped documents

Michael S. Brown, W. Brent Seales

Research output: Contribution to journalArticlepeer-review

121 Scopus citations

Abstract

We present a framework for acquiring and restoring images of warped documents. The purpose of our restoration is to create a planar representation of a once planar document that has undergone an arbitrary and unknown rigid deformation. To accomplish this restoration, our framework acquires and flattens the 3D shape of a warped document to determine a nonlinear image transform that can correct for image distortion caused by the document's shape. Our framework is designed for use in library and museum digitization efforts where very old and badly damaged manuscripts are imaged.

Original languageEnglish
Pages (from-to)1295-1306
Number of pages12
JournalIEEE Transactions on Pattern Analysis and Machine Intelligence
Volume26
Issue number10
DOIs
StatePublished - Oct 2004

Bibliographical note

Funding Information:
The authors gratefully acknowledge support for this work under the US National Science Foundation grant IIS-9817483 and the Hong Kong Research Grant Council grant

ASJC Scopus subject areas

  • Software
  • Computer Vision and Pattern Recognition
  • Computational Theory and Mathematics
  • Artificial Intelligence
  • Applied Mathematics

Fingerprint

Dive into the research topics of 'Image restoration of arbitrarily warped documents'. Together they form a unique fingerprint.

Cite this