Abstract

In this paper, I am presenting a preliminary step of document analysis and recognition .My aim is to extract textual areas, from gray scale document images by clustering collected areas into text and non-text areas basing on geometric and texture features proceeds in two steps. In the first step I retrieve interested areas via cumulative gradient considerations and secondly I classify them using certain geometric features, texture features and entropy criteria. Before that I have given a brief introduction to Digital image processing and relevance of my topic to DIP. Experiments are done on documents of different font types in English and I have also

…………….So on ..........(download any of the following links to get complete paper presentation in word document)