Solving the Binarization Challenges in Document Images Using OTSU Multilevel

摘要

This paper introduces a method for binarization of historical document images that suffer fromnon-uniform background, faint text, low contrast, stain, bleed-through, or shadow challenges.The proposed method adaptively detects the non-uniform background in the document imageand eliminates it. Areas that contain missing text are adaptively identified and reprocessedseparately. Stain and bleed-through objects are found depending on stroke width and locallybinarized. Shadow is detected based on the image contrast. Otsu multilevel is applied forbinarization. DIBCO series is used for testing.