Text-Color-Independent Binarization for Degraded Document Image Based on MAP-MRF Approach
Hideaki Orii, Hideaki Kawano, Hiroshi Maeda, Norikazu Ikoma · IEICE Transactions on Fundamentals of Electronics Communications and Computer Sciences · 2011
We propose a novel background and foreground estimation algorithm in MAP-MRF approach for binarization of degraded document image. In the proposed algorithm, an assumption that background whiteness and foreground blackness is not employed differently from the conventional algorithm, and we employ character's irregularities based on local statistics. This makes the method possible to apply to the image with various colored characters, ex. outlined characters by colored background. The effectiveness and the validity are shown by applying the proposed method to various degraded document images.