Open AccessObject- based Image Classification using Ant Colony Optimization and Fuzzy Logic
Author(s) -
Subba Rao K,
Sambasiva Rao N,
P. Sammulal
Publication year - 2019
Publication title -
international journal of innovative technology and exploring engineering
Language(s) - English
Resource type - Journals
ISSN - 2278-3075
DOI - 10.35940/ijitee.b6283.129219
Subject(s) - pixel , artificial intelligence , pattern recognition (psychology) , fuzzy logic , computer science , computer vision , ambiguity , image processing , image (mathematics) , mathematics , programming language
Image analysis enables to get meaningful information from a digital image by applying the image processing techniques. During the process of extracting of this meaningful information number of challenges needs to be addressed for a high-resolution image. These high-resolution images which contain minimum of 300 pixels per inch are also known as ‘coarse’ images. One common problem of coarse image is it combines the spectral properties of intermixed pixels. This nature of coarse image will lead to ambiguity in grouping the pixels into clusters which are in turn constitute to different objects in the input image. To reduce this ambiguity in classifying or grouping the pixels, Ant Colony Optimization and Fuzzy Logic which is a Hybrid classification technique is proposed. The ACO has solved many classification problems. In this paper ACO and Fuzzy based to group the pixels into meaningful groups for coarse image and results are compared with various other unsupervised classification methods such as ISODATA and K-means