Feature Extraction (Image Compression) of Printed Gujarati and Amharic Letters Using Discrete Wavelet Transform
Archit Ajay Yajnik, Dharmendra Singh · SAMRIDDHI A Journal of Physical Sciences Engineering and Technology · 2010
This paper demonstrates an application of discrete wavelet transform as a feature extractor as applicable to natural language processing. The procedure discussed in this paper extracts important features of the printed characters of Gujarati and Amharic scripts and shows Image compression capabilities of discrete wavelets. Procedure prescribed in this paper compresses the original image to 75% without disturbing much the original image.