Monitoring And Prediction Of Natural Disaster Using Multi-kernel Inception Vggnet-16 Classification
Authors: Seelam Ch. Vijaya Bhaskar, Dr. Anitha S
DOI: 10.87349/JBUPT/281013
Page No: 17-28
Abstract
The whole world is revealed to the disaster that causes severe damage to the life, infrastructure, and injury to the public. Human capabilities can be extended to predict and manage natural disasters with the help of the Internet of things and technology. The main objective of this paper is to design and implement sensor-based natural disaster monitoring and predicting methods in the regions of Krishna Godavari region of Andhra Pradesh state which helps to alert peoples at the correct time and provide the right decision. Initially, the datas from the cloud storage are acquired in the form of satellite images that are associated to the natural disasters. The collected data is preprocessed and then the feature extraction is employed using Matrix-based Fisher discriminant analysis and feature selection is performed by Adaptive Entropy-based PCA for Feature Selection method. The best resultant features were then optimized using the Heuristic Echolocation BAT optimization algorithm. The optimized output can be given as an input for the process of classification. The classification process is enhanced using Multikernel Inception VGGNet-16 classification. The output after classification is subjected to the alert system which is responsible for offering warning messages and for the detection and prediction of the natural disasters in order to make the right decisions. The performance analysis and the estimation of accuracy are done to prove the effectiveness of the proposed technique. The recommended technique proves to be favorable in monitoring and predicting natural disasters.
