Image dewarping and text extraction from mobile captured distinct documents

AbstractCamera Based Document Analysis (CBDA) is an emerging field in computer vision and pattern recognition. In recent days, cameras are moulded with several items of additional equipment. Thus, they play a vital role in the replacement of scanners with hand-held imaging devices (HIDs) like digital cameras, mobile phones and gaming devices. Warping is a common appearance in camera captured document images. It is the primary factor that makes such kind of document images hard to be recognized. Therefore it is necessary to restore warped document images before recognition. This paper presents a novel methodology to dewarp text from camera-captured warped images by co-ordinate transform model. Further, the extracted text from the warped and other document images will be recognized by maintaining a suitable database of all letters and numbers and converted into an editable form such as Notepad or as a MS Word document. The experimental results will be evaluated using a novel method to authenticate the methodology designed. Extensive experiments have been carried out on distinct documents and results are tabulated. Experimental results show the effectiveness of the proposed method