Goal-Oriented Rectification of Camera-Based Document Images

Abstract

Abstract-Document digitization with either flatbed scanners or camera-based systems results in document images which often suffer from warping and perspective distortions that deteriorate the performance of current OCR approaches. In this paper, we present a goal-oriented rectification methodology to compensate for undesirable document image distortions aiming to improve the OCR result. Our approach relies upon a coarse-to-fine strategy. First, a coarse rectification is accomplished with the aid of a computationally low cost transformation which addresses the projection of a curved surface to a 2-D rectangular area. The projection of the curved surface on the plane is guided only by the textual content's appearance in the document image while incorporating a transformation which does not depend on specific model primitives or camera setup parameters. Second, pose normalization is applied on the word level aiming to restore all the local distortions of the document image. Experimental results on various document images with a variety of distortions demonstrate the robustness and effectiveness of the proposed rectification methodology using a consistent evaluation methodology that encounters OCR accuracy and a newly introduced measure using a semi-automatic procedure.Index Terms-Document image analysis, document image processing, document image rectification, image dewarping.

Effective geometric restoration of distorted historical document for large‐scale digitisation

“…Compared to our previous Prima algorithm, GRLSD system with optimization of generated mesh has improved moderately the overall accuracy between 2%-5%. Particularly on some images (ID: 7,12,14,15), GRLSD system has apparent enhancement than Prima algorithm. Only on image 21, GRLS system performs slightly lower accuracy than Prima solution.…”

“…The accuracy of each document is estimated by the average baseline straightness of the original and the corrected image is calculated according to equation (2). The results of three additional geometric correction methods are compared: our initial arbitrary dewarping approach by Prima group [2], a state-of-the-art page-curl correction method designed for IMPACT by NCSR [12] and the leading commercial product Book Restorer™ [28].…”

“…Page curl correction approaches normally build a geometrical model by analysing image features [3][4][5][6][7][8] or using scanning specialised scanning hardware [9][10][11][12]. Page curl usually causes consistent global distortions on document images.…”

“…An unsupervised mesh generation approach is introduced to enhance the accuracy of baseline extraction, with a capability of robustly correcting arbitrary warping effect. In comparison to the current leading geometric restoration method [12] and industry standard commercial system [27], the proposed approach has an improved performance.…”

Precise Vehicle Speed Measurement Based on a Hierarchical Homographic Transform Estimation for Law Enforcement Applications

“…In literature, the homographic transformation estimation from text images is used in document analysis [15], [23] and natural scene text recognition systems [5], [14], [18], [19]. Documents used in document analysis systems usually have multiple lines and these lines are used to extract the homographic transformation.…”