Paper Title:Image Classification

Abstract

Digital image processing is a n method to perform some operations on digital images using the computer algorithms. Processing of images varies from one application to another, i.e. starting from image preprocessing (enhancement, restoration), segmentation, compression, rendering etc., on different types of images . Performance of image processing algorithms strongly depends on the type of the input image and its quality. This paper provides an algorithm to classify the images based on their hom ogeneity such as more homogeneous images, less ho mogeneous image, non homogeneous images and complex images, using the Interquartile Range (IQR). Performance is e valuated by experimental results obtained on various images taken fro m the standard database set.


Keywords:Interquartile range; classificatio n; Homogenous; Texture; blocks; Histogram;