DocumentCode :
2658652
Title :
Segmentation of Iranian bank checks
Author :
Alirezaee, Shahpour ; Ahmadi, Majid ; Alijamat, Ali ; Alahiari, Mohamad Ali
Author_Institution :
Abhar Branch, Islamic Azad Univ., Abhar
fYear :
2008
fDate :
23-24 Dec. 2008
Firstpage :
460
Lastpage :
462
Abstract :
This paper aims to provide a document analysis algorithm for the Iranian bank checks. The main idea is based on the connected component concept. The proposed algorithm uses the connected component properties to segment unconstrained Iranian bank checks and prepares those for OCR application. To evaluate the performance of the algorithm, it has been tested on a set of 500 Iranian bank checks which collected from 8 types of Iranian bank checks. The algorithm has a good success on document restoration and segmentation. Numerical results indicate that the proposed algorithm has 94.04% accuracy on segmentation of 5 fields of the Iranian bank checks.
Keywords :
cheque processing; document image processing; image restoration; image segmentation; optical character recognition; Iranian bank check; connected component concept; document analysis algorithm; document restoration; document segmentation; optical character recognition; Algorithm design and analysis; Character recognition; Gray-scale; Handwriting recognition; Image restoration; Image segmentation; Morphology; Smoothing methods; Text analysis; Text recognition; Connected Component; Document Analysis; Segmentation;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Multitopic Conference, 2008. INMIC 2008. IEEE International
Conference_Location :
Karachi
Print_ISBN :
978-1-4244-2823-6
Electronic_ISBN :
978-1-4244-2824-3
Type :
conf
DOI :
10.1109/INMIC.2008.4777782
Filename :
4777782
Link To Document :
بازگشت