A partition approach for the restoration of camera images of planar and curled document

摘要

As camera resolution increases, high-speed non-contact text capture through a digital camera is opening up a new channel for text capture and understanding. Unfortunately, the captured document images are normally coupled with the perspective and geometric distortions that cannot be handled by the existing optical character recognition (OCR) systems. In this paper, we propose a new technique, which is capable of removing the perspective and geometric distortions, and reconstructing the fronto-parallel view of text with a single document image. Different from reported approaches in the literature, the restoration of the distorted camera documents is carried out through the image partition, which divides the documents into multiple small image patches where text can be approximated to lie on a planar surface. The global distortion is thus corrected through the local rectification of the partitioned image patches one by one. Experimental results show that the proposed method is fast and easy for implementation.