Ternary Entropy-based Binarization of Degraded Document Images using Morphological Operators.

The purpose of this project is to improve and binarize an image of degraded document for text detection. The project is developed using MATLAB. The following are features of this project: A contrast adjustment is performed on the image in the first step. Shannon entropy-based ternarization thresholding is performed on the second stage. The near text values are relabelled. I have implemented morphological operations in the same. Also applied shrink and swell filters.

Note: This project, the files and / or the code should be used only as a reference and should not be used as such or after modification. The author is not responsible for any loss or harm or other unexpected events caused by accessing, using or running this project or any part of it.