Due to excessive human interference, the natural equilibrium has been disrupted, leading to an increase in natural disasters. This disruption can potentially result in the loss of human technology and hinder technological development within affected countries. To diminish the impact of disasters on human lives and natural resources, it is vital to locate and react to such events as soon as possible. Here comes the significant of remote sensing, especially microwave remote sensing technologies like Synthetic Aperture Radar (SAR), which can pierce harmful conditions such as heavy rain, strong winds, thunderstorms, and even explosions. In this systematic analysis, we have employed VV and VH channels of Synthetic Aperture Radar (SAR) images and the world cover map of the European Space Agency (ESA). A semantic segmentation is carried out using deep neural network. The flood prediction accuracy of 85.79% is obtained using VV, VH and ESA world cover map features, which is much better when compared to the results obtained using VV and VH bands. This result highlights the significant contribution of combining SAR imagery with the ESA World Cover map for more accurate flood prediction. Our findings emphasize the importance of integrating multiple data sources, such as ESA World Cover maps and SAR imagery, to enhance the reliability of flood prediction models, ultimately contributing to better disaster preparedness and mitigation strategies.