Metric Rectification to Estimate the Aspect Ratio of Camera-Captured Document Images

Document images from mobile phone cameras and digital cameras suffer from geometric distortion; therefore distortion correction is desired for better character recognition. We propose a method to calculate the aspect ratio of planar documents using 3D perspective projection of quadrangles without using vanishing points. The proposed method is based on estimation of direction of a pair of parallel lines from their projected image. We verify the accuracy of our method from experimental results at various viewing angles. Our contribution can be applied to increase the character recognition rate by compensating for distorted documents with the accurate aspect ratio.

[1]  Tieniu Tan,et al.  Robust automated ground plane rectification based on moving vehicles for traffic scene surveillance , 2008, 2008 15th IEEE International Conference on Image Processing.

[2]  Peter F. Sturm,et al.  Using geometric constraints through parallelepipeds for calibration and 3D modeling , 2005, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3]  Yisong Chen,et al.  Planar metric rectification by algebraically estimating the image of the absolute conic , 2004, ICPR 2004.

[4]  Shinichiro Omachi,et al.  Layout-free dewarping of planar document images , 2009, Electronic Imaging.

[5]  C. V. Jawahar,et al.  Perspective Correction Methods for Camera-Based Document Analysis , 2005 .

[6]  David S. Doermann,et al.  Geometric Rectification of Camera-Captured Document Images , 2008, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[7]  Olivier D. Faugeras,et al.  Determination of Camera Location from 2-D to 3-D Line and Point Correspondences , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  Andrew Zisserman,et al.  Multiple View Geometry in Computer Vision (2nd ed) , 2003 .

[9]  K. Kanatani.  3D recovery of polyhedra by rectangularity heuristics , 1989, International Workshop on Industrial Applications of Machine Intelligence and Vision,.

[10]  Andrew Zisserman,et al.  Metric rectification for perspective images of planes , 1998, Proceedings. 1998 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (Cat. No.98CB36231).

[11]  Zhengyou Zhang,et al.  A Flexible New Technique for Camera Calibration , 2000, IEEE Trans. Pattern Anal. Mach. Intell..

[12]  Bernhard P. Wrobel,et al.  Multiple View Geometry in Computer Vision , 2001 .

[13]  L. Lucchese.  Closed-form pose estimation from metric rectification of coplanar points , 2006 .