Novel classification approach for predicting the center of tropical cyclone tc

Abstract

Over the past centuries, the impact of the cyclone on human lives newlineand property is huge. Depending on the geographical origin, most cyclone is newlinecalled Tropical cyclone since it forms over tropical seas. Since the diameter of newlinetropical cyclones varies between 100 and 2000 km, it is causing huge damage. newlineThe winds will be whirling around the central eye which is to be found to newlineidentify the exact location of the tropical cyclone. A method called Content- newlineBased Image Retrieval (CBIR) can help in searching and retrieving the image newlinefrom a vast database. Briefly, Image retrieval is a technique for searching a newlinebig image library for the most visually comparable images to a given query newlineimage. The main benefit of this method is that it requires very little human newlineinteraction. In our research, we aim to identify the eye of the cyclone to locate newlinethe cyclone position precisely from the database. The most difficult aspect of newlinethis procedure is retrieving the needed images from a vast database with the newlinehighest degree of precision and in the shortest amount of time possible. As a newlineresult, an effective image retrieval system is necessary to provide a userfriendly newlinesolution for retrieving relevant images from a big database in a short newlineamount of time with high accuracy. In this study, we enhance the design of newlinethe Content Based Image Retrieval (CBIR) system as the use of the CBIR newlinesystem automatically improves the resultant images with high quality and we newlinefurther use different image-processing techniques to process the raw cyclone newlineimage. With the growing platform of satellite images, image retrieval is an newlineintriguing field for scholars to investigate. The suggested work is presented newlineusing the MATLAB programming language. However, when estimating the newlineiv suggested classifier, many factors are taken into account. In this research, newlinemany current classifiers such as Neural Network (NN), Convolution Neural newlineNetwork-Whale Optimization Classifier (CNN-WOC), and Convolution newlineNeural Network-Elevated Whale Optimization Classifier (CNN-EWOC) are newlineutilized to