Satellite Image-Based Flood Monitoring and Prediction using Multi Verse Optimization Algorithm with Support Vector Machine

One of the deadliest and most catastrophic natural disasters is floods which cause a significant number of damages in infrastructures and agricultural land. Real-time flood management systems locate flooded regions and initiate relief operations with faster and efficient way. Therefore, this research presented Multi-Verse Optimization Algorithm-Support Vector Machine (MVOA-SVM) is presented for flood detection by utilizing satellite images. Initially, the collected images is employed to get high spatial resolutions in flood region. Then noise reduction and orthorectification are used to improve the superiority of images. After that feature selection using Landmark detection is then utilized to select the high-quality image with effective features and provide accurate outcomes. In order to simplify flood monitoring systems connected to remote sensing networks, the MVOA-SVM system is efficiently proposed for monitoring the land devastated by floods. When compared with existing models such as Random Forest (RF), Support Vector Machine (SVM), and Artificial Neural Network (ANN) the suggested MVOA-SVM method performed better results in terms of accuracy of 99.95%, precision of 99.74%, sensitivity of 99.26% and specificity of 99.61% respectively.