RECOGNITION AND DATA EXTRACTION OF FORM DOCUMENTS BASED ON 3 TYPES OFLINE SEGMENTS

Almost all form documents contain line segments. In this paper, we pro pose an efficient method to recognize the form document that contains at least one line segment. Our method is based on an efficient represe ntation model of the form. The representation model uses three types o f line segments to represent a form. All line segments are normalized and sorted after they were extracted. The normalization and sorting no t only solve the form scaling problem but also provide an unified and efficient way of matching between forms. To make the recognition metho d more robust, a fuzzy matching is used. Using the representation mode l, when recognizing a skew form, only the line segments and the data f ields instead of the whole form image need to be rotated. Experimental results show the effectiveness and the efficiency of the method. (C) 1998 Pattern Recognition Society. Published by Elsevier Science Ltd. A ll rights reserved.