This paper presents a new hybrid approach for the binarization and enhancement of Historical Manuscript. This paper deals with degradations which occur due to shadows, non-uniform illumination, low contrast and strain. We follow two distinct method of Binarization with a pre-processing procedure using a adaptive Wiener filter, a rough estimation of foreground regions and a background surface calculation by interpolating neighboring background intensities. Further logical anding of the calculated background surface with compliment of second method result, performing final thresholding and post-processing in order to improve the quality of text regions. After extensive experiments, our method demonstrated superior performance against some wellknown techniques on numerous degraded document images as well as on Historical Manuscript in both manners qualitatively and quantitatively.
Goyal, Rajan and Kaur, Amandeep
"HYBRID BINARIZTION TECHNIQUE FOR HISTORICAL MANUSCRIPTS,"
International Journal of Image Processing and Vision Science: Vol. 1
, Article 2.
Available at: https://www.interscience.in/ijipvs/vol1/iss2/2