Anchor point selection by KL-divergence

Selecting anchor points for the identification of scanned documents can be an effective and quick means of identifying unknown documents. Here, we discuss and compare some strategies for classification of scanned forms using anchor points and show experiments indicating that a robust system can be built with only a few training examples.

[1]  Xiaofan Lin.  DRR research beyond commercial off-the-shelf OCR software: a survey , 2005, IS&T/SPIE Electronic Imaging.

[2]  Shiri Gordon,et al.  An efficient image similarity measure based on approximations of KL-divergence between two gaussian mixtures , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[3]  Xiaofan Lin.  Drr research beyond COTS OCR software : A survey , 2005 .

[4]  Roberto Brunelli,et al.  Template Matching Techniques in Computer Vision: Theory and Practice , 2009 .

[5]  Prateek Sarkar.  Learning Image Anchor Templates for Document Classification and Data Extraction , 2010, 2010 20th International Conference on Pattern Recognition.

[6]  David J. Hand,et al.  Measuring classifier performance: a coherent alternative to the area under the ROC curve , 2009, Machine Learning.

[7]  Nenghai Yu,et al.  Efficient near-duplicate image detection by learning from examples , 2008, 2008 IEEE International Conference on Multimedia and Expo.

[8]  Michael D. Garris,et al.  Intelligent system for reading handwriting on forms , 1998, Proceedings of the Thirty-First Hawaii International Conference on System Sciences.

[9]  Roberto Brunelli,et al.  Advanced , 1980 .