A DIFFERENTIAL-PROCESSING EXTRACTION APPROACH TO TEXT AND IMAGE SEGMENTATION

To efficiently store the information found in paper documents, text an d non-text regions need to be separated. Non-text regions include half -tone photographs and line diagrams. The text regions can be converted (via an optical character reader) to a computer-searchable form, and the non-text regions can be extracted and preserved in compressed form using image-compression algorithms. In this paper, an effective syste m for automatically segmenting a document image into regions of text a nd non-text is proposed. The system first performs an adaptive thresho lding to obtain a binarized image. Subsequently the binarized image is smeared using a run-length differential algorithm. The smeared image is then subjected to a text characteristic filter to remove error smea ring of non-text regions. Next, baseline cumulative blocking is used t o rectangularize the smeared region. Finally, a text block growing alg orithm is used to block out a text sentence. The recognition of text i s carried out on a text sentence basis.