Flood is not a new disaster that we face nowadays in every part of the world. It is sudden, fast, and the impact is beyond the imagination. Its frequency is increasing day by day. Although we can't avoid this natural disaster, We should manage it properly. For that, image detection has a great role and should find the best classifier to detect it. The classifiers we use are knearest neighbors, Logistic Regression, Support Vector Classifier, Decision Tree, and Random Forest machine learning algorithms. By learning through each algorithm we found the best among them. The accuracy obtained by learning each algorithm on our trained model is quite different and we found out the best. First, we prepared the image dataset which includes remote dataset and satellite images. Second, we passed the dataset to each classifier and obtained the variant accuracies. Best results are produced in each method. The classifier which gives the best can be taken for the early prediction of the flood. By using new technologies to manage the flood will help us with evacuation faster and take care of people who are affected. Flood prediction has done here using history rainfall data so that we just predicted the chance. Detection is done mainly with high accuracy and the accuracy of each classifier is shown. Also the image tested result shown.

Keywords : Flood Detection ; Accuracy ; Training ; Convolutional Neural Network ;Logistic Regression ; KNearest Neighbor ; Naive Bayes Classifier ; Support Vector Machine ; Synthetic Minority Oversampling Technique