Call For Paper Volume:4 Issue:8 Aug'2017 |

Document Image Binarization Technique For Enhancement of Degraded Historical Document Images.

Publication Date : 20/04/2016



Author(s) :

Manish Deelipkumar Wagh , Mayur Yashwant Bachhav , Vijay Balasaheb Gare.


Volume/Issue :
Volume 3
,
Issue 4
(04 - 2016)



Abstract :

Segmenting text from image of rough, degraded historical document is difficult task because of higher inter variation and intra variation between foreground text and document background of different document images. This can be solved with the help of image binarization technique. We can make use of this technique to clear the historical document, making them usable for further processing. As the image of old historical documents are often in degrade form it is hard to retrieve and understand what is written on them. So it is necessary to find solution to this problem, by taking picture and binarizing it with suitable technique we can able to understand the text so here we developed a image binarization technique. Here we provide new segmentation algorithm in which each pixel has its own threshold value. In old system the contrast image is used as preprocessing step but in this technique gray scale image is used. The pixels are categorized as foreground or background pixels depending upon their comparison with threshold valued. Pixel value is compared to the threshold value. The technique is known as window thresholding .We are doing work on window of size p*q and from each window text stroke of each pixel are extracted .Again local threshold is used to segment the document text that is estimated based on intensity of detected text stroke of edge pixel in local window.


No. of Downloads :

8


Indexing

Web Design MymensinghPremium WordPress ThemesWeb Development

Document Image Binarization Technique For Enhancement of Degraded Historical Document Images.

April 13, 2016