CSDL Home IEEE Transactions on Pattern Analysis & Machine Intelligence 2008 vol.30 Issue No.04 - April
Issue No.04 - April (2008 vol.30)
Daniel DeMenthon , IEEE
David Doermann , IEEE
Compared to typical scanners, handheld cameras offer convenient, flexible, portable, and non-contactimage capture, which enables many new applications and breathes new life into existing ones. However,camera-captured documents may suffer from distortions caused by non-planar document shape andperspective projection, which lead to failure of current OCR technologies. We present a geometricrectification framework for restoring the frontal-flat view of a document from a single camera-capturedimage. Our approach estimates 3D document shape from texture flow information obtained directlyfrom the image without requiring additional 3D/metric data or prior camera calibration. Our frameworkprovides a unified solution for both planar and curved documents and can be applied in many, especiallymobile, camera-based document analysis applications. Experiments show that our method producesresults that are significantly more OCR compatible than the original images.
Camera-based OCR, image rectification, shape estimation, texture flow analysis.
Daniel DeMenthon, David Doermann, "Geometric Rectification of Camera-Captured Document Images", IEEE Transactions on Pattern Analysis & Machine Intelligence, vol.30, no. 4, pp. 591-605, April 2008, doi:10.1109/TPAMI.2007.70724