Search

This paper presents a support vector machine (SVM) based method for degraded historical document image binarization. Given a degraded historical document image, the proposed method first segments the image into w  ×  w regions and implements a local contrast enhancement in each image block. We then use a SVM to select an optimal global threshold for binarization of each image block. Finally, the entire image is further binarized by a locally adaptive thresholding method. The proposed method has been evaluated over the recent Document Image Binarization Competition (DIBCO) datasets. The experimental results show that our proposed method outperforms other state-of-the-art techniques in terms of F-measure, NRM, DRD, and MPM. [ABSTRACT FROM AUTHOR]

Historical document restoration has always been a challenging need for information acquisition and conversion to a stable and portable format. Document image binarization, is an important pre-processing phase in document image analysis and processing of such historical documents. It has never been an easy task accounting to the difficulty in choosing a threshold value when the document image suffers from noises due present in the original document due to its degradation, ageing of the paper, other unwanted impurities, background images, bleed-through, and so on. Inconsistency in the darkness of the letters in handwritten documents adds to the difficulty. Though there are many algorithms that have been proposed for document image binarization, there has not been an effective algorithm which can efficiently binarize all kind of documents. In this paper we introduce a novel method to perform document image binarization using bit-plane slicing. As per this approach, the 8 bit planes of the gray image are extracted and processed separately, at the end of which, the results are combined to give the final binarized output. This method works well for all kinds of historical documents except those with both, less noise and low contrast. [ABSTRACT FROM AUTHOR]