Table-form Extraction with Artefact Removal

In this paper we present a novel methodology to recognize the layout structure ofhandwritten filled table-forms. recognition methodology includes locating line intersections,correcting wrong intersections produced by what we call artefacts (overlapping data, brokensegments and smudges), extracting correct table-form cells and using as little previous tableformknowledge as possible. to improve layout structure recognition, a novel artefactidentification and deletion method is also proposed. to evaluate the effectiveness of themethodology, a database composed of 350 handwritten filled table-form images damaged bydifferent types of artefacts was used. experiments show that the artefact identification methodimproves performance of the table-forms structure extractor that reached a success rate of 85%.