Automated document image preprocessing management utilizinggrey-scale image analysis and neural network classification

摘要

This paper describes a new approach for preprocessing grey-scaledocument images. The grey-scale image is first partitioned and analyzedusing a set of image features to extract the quality and contentsinformation in the given entity. Then, this data is classified using aneural network classifier to find and prioritize the need for imageenhancement, if any. The classification information guides a specialfilter bank and their parameters to perform prefiltering operation todocument image. In filtering, a control method is used to assure that nooverfiltering occurs. The system behaviour and experiments are presentedshowing the performance of the developed techniques with tests performedwith three different document databases