Transformation of arc-form-text to linear-form-text suitable for OCR

摘要

Artistic text contents are quite common in several documents such as certificates, advertisements, historic documents. One such artistic structure is arc-form-text. If an OCR has to be employed, it is necessary to transform the text in arc-form to linear-form-text. In this paper, we present an approach to accomplish this. Since the peripheral of an arc-form-text approximates concentric ellipsoidal contours, the proposed model is based on ellipse drawing algorithm. The methodology sequences through three stages - first it involves segmentation of document containing arc-form-text and binding the top and bottom contours of the arc peripherals by concentric ellipsoidal enclosures, next transformation of arc-form to linear form takes place and finally tilt deformation present in the transformed characters is corrected. The proposed method is implemented on several upper semi-circular and arc-form text inputs, and the readability of the transformed text is analyzed with an OCR.