Global and Local Shape Prior for Variational Segmentation of Degraded Historical Characters

We propose a variational method for model based segmentation of highly degraded gray scale images of historical documents. Given a training set of characters (of a certain letter), we construct a small set of shape models that cover most of the training set’s shape variance. For each gray scale image of a respective degraded character, we construct a custom made shape prior using those fragments of the shape models that best fit the character’s boundaries. Therefore, we are not limited to any particular shape in the shape model set. Experiments show that our method achieves very accurate results in segmentation of highly degraded characters. When compared with manual segmentation, the average distance between the the boundaries of respective segmented characters was 0.8 pixels (the average size of the characters is 70 ∗ 70 pixels).

[1]  Junaed Sattar.  Snakes , Shapes and Gradient Vector Flow , 2022 .

[2]  Demetri Terzopoulos,et al.  Snakes: Active contour models , 2004, International Journal of Computer Vision.

[3]  Guillermo Sapiro,et al.  Geodesic Active Contours , 1995, International Journal of Computer Vision.

[4]  Tin Kam Ho,et al.  Enhancing degraded document images via bitmap clustering and averaging , 1997, Proceedings of the Fourth International Conference on Document Analysis and Recognition.

[5]  Michael Droettboom.  Correcting broken characters in the recognition of historical printed documents , 2003, 2003 Joint Conference on Digital Libraries, 2003. Proceedings..

[6]  William H. Offenhauser,et al.  Wild Boars as Hosts of Human-Pathogenic Anaplasma phagocytophilum Variants , 2012, Emerging infectious diseases.

[7]  Yunmei Chen,et al.  Using Prior Shapes in Geometric Active Contours in a Variational Framework , 2002, International Journal of Computer Vision.

[8]  J. Sethian,et al.  Fronts propagating with curvature-dependent speed: algorithms based on Hamilton-Jacobi formulations , 1988 .

[9]  Manuel Menezes de Oliveira Neto,et al.  Fast Digital Image Inpainting , 2001, VIIP.

[10]  O. Faugeras,et al.  Statistical shape influence in geodesic active contours , 2002, 5th IEEE EMBS International Summer School on Biomedical Imaging, 2002..

[11]  Ioannis Pratikakis,et al.  Adaptive degraded document image binarization , 2006, Pattern Recognit..

[12]  Nadia Bali,et al.  Automatic accurate broken character restoration for patrimonial documents , 2006, International Journal of Document Analysis and Recognition (IJDAR).

[13]  Itay Bar-Yosef.  Input sensitive thresholding for ancient Hebrew manuscript , 2005 .

[14]  Rachid Deriche,et al.  A Review of Statistical Approaches to Level Set Segmentation: Integrating Color, Texture, Motion and Shape , 2007, International Journal of Computer Vision.