Classification Techniques for Object Detection in Remote Sensing Images

Abstract

In the field of remote sensing, multispectral images contain the major part of information which can be processed to identify objects. Classification of satellite images is playing key role for various object recognition system. Satellite image classification needs selection of appropriate classification techniques based on the requirements. In the process of object identification different classification techniques used are SVM, k-NN, Naïve Bayes, Random Forest, Artificial Neural Network, Convolutional Neural Network, Recurrent Neural Network. In this paper different classification techniques have been reviewed and analyzed in terms of performance.

Keywords: Support Vector Machine (SVM), k-NN (k-Nearest Neighbor), Convolutional Neural Network (CNN)