Research Project

An image classification model by combining Convolutional Neural Networks and Support Vector Machine for land use and land cover scene classification. CNN namely, ResNet-152 was modified and used for deep feature extraction. A stack of fully-connected layers and SVM both were compared for scene classification. This project was implemented using Python programming language

Configuration_Manual.pdf :

This configuration manual helps the user's to understand and follow the steps involved in implementing the hybrid models used for land use and land cover scene classification. System configuration, development environment artefacts, process input and output as well as steps involved in experimentation are explained in this manual.