An Efficientbinarization Technique for Historical Document Images

Binarization is an important preprocessing step in several document image processing tasks. Binarization of historical document with poor contrast, strong noise, and non-uniform illumination is a challenging problem. A new binarization algorithm has been developed to address this problem. In this paper, we describe a new method which does not utilize statistical properties of the intensity histogram of a gray-scale image to determine a threshold. Our method is based on the “Divide and Conquer” strategy for calculating a threshold value which is based on the list of gray levels in that image. It uses low pass Weiner filter method as a preprocessing step to enhance the image by deblurring the image. It uses Median Filter method as a postprocessing step to reduce noise in the image. Our result shows that this new binarization method produce high quality binary image for historical document than any other global methods such as Otsu’s method.