Adaptive binarization of degraded documents
This paper presents a new adaptive binarization method for the degraded document images. Variable background, non-uniform illumination, ink bleed-though and blur caused by humidity are the addressed degradations. The proposed method has four steps: contrast analysis which calculates the local contrast threshold, cumulative histogram stretching to transform original image into a degradation free image, thresholding by computing global threshold and information recovery to recover lost foreground objects to fix broken and thin text. Evaluation has been done on three sets of images: ground-truth based evaluation using established measures, OCR based evaluation and evaluation on visual basis. The results were tested against eight well-known techniques referred in the literature.