Performance of Most Common Thresholding Techniques over a Generic Structured Document Classifier
Hamam Mokayed, Azlinah Mohamed · International journal of imaging and robotics · 2014
Developing an appropriate binarization method for an input image to the generic document classifier is a difficult problem. Typically, a human expert evaluates the binarized images according to his /her visual criteria. However, to conduct an objective evaluation, one needs to investigate how well the subsequent image analysis steps will perform on the binarized image. For the previous mentioned reason, a study on most common thresholding techniques, express their formulas, and evaluate their performance as a pre-step in a generic structured document classifier is conducted. This study will address challenges faced in binarization which affected quite critically the performance of the successive steps in classification. The most important challenge is the value of the Misclassification Error of a document which is referred to as Type I error or commonly called ME.