RESEARCH ON THE EXISTING IMAGE CLASSIFICATION METHODS

Abstract

Image classification is a topical issue in data science, which is applied in many areas. It is the process of getting classes of information from a multichannel bitmap using specific rules. There are three approaches to image classification - supervised, unsupervised and objectbased. Support Vector Machines, Artificial Neural Network, Decision Tree, Convolutional Neural
Network are the solution methods used for image classification. Choice of the classification approach depends on the way an analyst interacts with the computer. The whole process boils down to image stack acquisition, preprocessing and classification.