MACHINE LEARNING BASED IMAGE PROCESSING USING UNSUPERVISED APPROACH

Abstract

Enhancing the visual media for the purpose of better perception has been a research topic for years. It finds its secondary application in the recognition of objects, analysis of medical images accounting the astronomical data and so on. The disintegration of an image based on its meaningful components plays a key role in many image processing applications like filtering, interpolation, image enhancement, feature variation, etc. the solution to this vary from basic segmentation techniques to advanced methods like fuzzy logic and machine learning. Through this paper, we present a novel method of image processing using machine learning algorithms. We also conduct experiments with preliminary image processing techniques and provide comparable performance measures to illustrate the success of our approach.