In recent years, the population's safety and property have been seriously threatened by the rising frequency and severity of floods. Thus, in order to reduce any damage, it is imperative to identify flood-prone locations as soon as possible. Many methods are used to detect floods, but one particularly useful option is satellite imaging. Images captured by satellites, which are taken from a distance above the planet, are heavily noisy. Fuzzy clustering is applied together with a clustering technique to deal with this noise. In this paper, we provide a new picture fuzzy clustering method for scene segmentation that is aimed at identifying flooded regions in satellite images. Experiments validate the proposed approaches and show that it is possible to identify flood-prone sites in satellite data. In particular, the proposed method produces pretty obvious identification of flood photos and good clustering accuracy. These developments are crucial for preventing disasters, particularly in the area of flood detection, since they enable prompt decision-making and suggest mitigating measures to lessen the damage that floods do to property and life.