People’s lives and property are put in jeopardy by extreme weather occurrences such as floods, but key infrastructures, which must work even under subpar conditions, are also severely damaged. There have been a number of attempts to use deep learning algorithms to map the urban flood regions. In order to determine how vulnerable an area is to catastrophic flooding, it is critical to evaluate the flood-prone zones. Using Sentinel satellite photos, along with deep learning model, can help to map flood-prone areas and assess their vulnerability. Flood hazard mapping was evaluated using a Convolutional Neural Network (CNN) model. The response variable was a flood inventory of Chennai Corporation’s flooded and non-flooded areas, while the predictor variables were flood-affecting factors. The flooded areas were then randomly divided into sections for the purpose of constructing flood models and testing them. According to the training and testing classification accuracy, the models were evaluated on their ability to predict outcomes. Flood protection and risk assessment might benefit from the use of the model. As a result, if a flood inventory is available, this method might be used to create a flood danger map for metropolitan areas.