Efficient Binarization of Degraded Document Image

Abstract: Image binarization is the method of separation of pixel values into dual collections, black as background and white as foreground. Thresholding has found to be a well-known technique used for binarization of document images. Thresholding is further divide into the global and local thresholding technique. In document with uniform contrast delivery of background and foreground, global thresholding is has found to be best technique. In degraded documents, where extensive background noise or difference in contrast and brightness exists i.e. there exists many pixels that cannot be effortlessly categorized as foreground or background. In such cases, local thresholding has significant over available techniques. The main objective of this paper is to evaluate the different image binarization techniques to find the gaps in existing techniques.

Keywords: Binarization; local thresholding; global thresholding; binary image.