DocumentCode :
2848480
Title :
Study on Template-Based Coding Method of Binary OCR Table Image
Author :
Guan, Zhengqiang ; Yang, Bo ; Li, Pengfei
Author_Institution :
Office of Acad. Affairs, Chongqing Univ. of Sci. & Technol., Chongqing, China
Volume :
2
fYear :
2010
fDate :
13-14 Oct. 2010
Firstpage :
498
Lastpage :
501
Abstract :
The binary image is commonly found with a lower compression ratio in lossless coding, for which a template-based coding method was employed in the coding of the OCR table images. Specifically, a template image was first established in the same context of the table image, followed by segmentation of the table image. Secondly, the deviation between the table image and the template image was used for rotation and translation to eliminate post-synthesis mismatch. Finally, adaptive arithmetic coding was introduced in the coding of the foreground image in the table image. In decoding, the template image and the foreground image are required to be synthesized. Experiments proved that the method used for the coding of binary OCR table images may have a higher compression ratio than the common coding method.
Keywords :
image coding; optical character recognition; adaptive arithmetic coding; binary OCR table image; lossless coding; post-synthesis mismatch; template image decoding; template-based coding method; Channel coding; Context; Context modeling; Image coding; Image segmentation; Object oriented modeling; Optical character recognition software; OCR table image; lossy coding; template coding;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent System Design and Engineering Application (ISDEA), 2010 International Conference on
Conference_Location :
Changsha
Print_ISBN :
978-1-4244-8333-4
Type :
conf
DOI :
10.1109/ISDEA.2010.202
Filename :
5743469
Link To Document :
بازگشت