Document Image Dewarping Based on Text Line Detection and Surface Modeling (RESEARCH NOTE)
Maryam Shamgholi · International Journal of Engineering · 2014
Document images produced by scanner or digital camera, usually suffer from geometric and photometric distortions. Both of them deteriorate the performance of OCR systems. In this paper, we present a novel method to compensate for undesirable geometric distortions aiming to improve OCR results. Our methodology is based on finding text lines by dynamic local connectivity map and then applying a low cost transformation to project curved area to 2-D rectangular area. We evaluate the performance of the proposed methods in combination with three participating methods on the public DFKI-I dataset (CBDAR 2007 dewarping contest), which contains camera-captured document images. Experimental results indicate the effectiveness and superiority of the proposed method.