BINARIZATION TECHNIQUE FOR DEGRADED DOCUMENT IMAGE

Computer Engineering

Now a days we can easily store, duplicate and backup our data which is in digital form. But sometimes the old documents may contain important information .They may get degraded after long span of time. Many of these degraded documents may have their foreground mixed with background. To read text from such degraded documents is very challenging. We present binarized documentation technique to segment text from such badly degraded documents. In the presented technique first we generate adaptive image contrast map is by giving degraded document images as input later we detect the text stroke edge pixel. Text of the document is segmented by using local threshold estimation then further the post processing is applied to improve binarization quality of document. This technique is effective, easy and involves minimum parameters. The output of this technique will produce a clear and binarized image. This technique will be presented on android platform and will be easily available to user as android application.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.