DocumentCode
459014
Title
Field Extraction Based on Two-level Regulated HMT in Auto Form Processing
Author
He, Xiuling ; Yang, Yang ; Chen, Zengzhao ; Yu, Ying ; Dong, Cailin
Author_Institution
Inf. Eng. Sch., Univ. of Sci. & Technol., Beijing
Volume
2
fYear
2006
fDate
16-18 Oct. 2006
Firstpage
716
Lastpage
719
Abstract
Data fields in a form contain invaluable information so that field data extraction is important in form processing. Most conventional data extraction methods depend on the line frames. However there are many types of forms such as table-form and non table-form that are without line frames and popularly used in our daily lives as well. A new field locating method using priori knowledge is proposed. It is based on two-level regulated hit or miss transform features of form background. The feasibility of the method is proved theoretically. Experiments and application show that it is several dozen times faster than ordinary regulated hit or miss transform and the correct rate is satisfied with real applications
Keywords
business forms; document image processing; feature extraction; auto form processing; field data extraction; field locating method; hit-and-miss transform; Computer science; Data engineering; Data mining; Image segmentation; Labeling; Mathematics; Morphological operations; Noise reduction; Noise robustness; Statistics;
fLanguage
English
Publisher
ieee
Conference_Titel
Intelligent Systems Design and Applications, 2006. ISDA '06. Sixth International Conference on
Conference_Location
Jinan
Print_ISBN
0-7695-2528-8
Type
conf
DOI
10.1109/ISDA.2006.253699
Filename
4021751
Link To Document