A Novel Degraded Document Image Binarazation by using Local Thresholding Segmentation

The proposed binarization is a scheme of parting a image pixel values into two classes black as foreground and white pixels as background then the thresholding is found for well known scheme for document image binarization. In this proposed work for the decomposition of both global and local thresholding this basic thresholding value we can use further. Here the global thresholding scheme is effectively improve the uniform contrast delivery to both foreground and background of the selected degraded document image. In some cases this global thresholding is not effortlessly categorized noise or difference of the either background or foreground. To overcome this problem the local threshold segmentation is available techniques. The scope of this paper is to binarization of degraded document images and to evaluate the different datasets.