Region-based document image denoising

Qing-Wen Zhou; Kai Wang; Hong-Jiang You; Qing-Ren Wang

doi:10.1117/12.866973

20 August 2010 Region-based document image denoising

Qing-Wen Zhou, Kai Wang, Hong-Jiang You, Qing-Ren Wang

Proceedings Volume 7820, International Conference on Image Processing and Pattern Recognition in Industrial Engineering; 78202G (2010) https://doi.org/10.1117/12.866973
Event: International Conference on Image Processing and Pattern Recognition in Industrial Engineering, 2010, Xi'an, China

Abstract

Traditional image de-noising methods mainly focus on the global effect, but the noise in document images tends to gather in certain local areas. So global de-noising processes will inevitably affect the recognition rate. Region based image de-noising uses pixel statistical information of local regions to separate noise and non-noise regions. And de-noising only applies on these noise regions instead of the entire image. So the deficiency of traditional method can be overcome. Test result on UNLV with 11176 samples shows that the average recognition rate rises from 94.44% to 94.85% by using this method.

Citation Download Citation

Qing-Wen Zhou, Kai Wang, Hong-Jiang You, and Qing-Ren Wang "Region-based document image denoising", Proc. SPIE 7820, International Conference on Image Processing and Pattern Recognition in Industrial Engineering, 78202G (20 August 2010); https://doi.org/10.1117/12.866973

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available