AL-Rafidain Journal of Computer Sciences and Mathematics

Abstract

Due to the fast and high development in computer technology and the reflection of this development on digital images ,many image processing algorithms became in need to initialization steps for the image before starting the actual operations of the algorithms and the program. Image quantization is one of the important operations in image processing field and it is the first step in  many digital image applications.
This technique is based on taking the best colors  from the original image and produce a new image with a new quality with less colors and with small error ratio. There are many image quantization methods ,but this research focuses on studying three different methods of image quantization and compares  them. These methods are:

Quantization by mask
Uniform Quantization
Half toning Quantization


After programming the methods we could reach a high degree of clarity in image after reducing  its color. The research was applied on different types of images to find the best method for each image  with small error ratio and that depends on the contents and on color distributions in the image .(Matlab7) was used for programming these methods.