In this work, we aim to develop a simple and fast algorithm using conventional methods in images segmentation for the automatic detection and extraction of upwelling areas, in the coastal region of Morocco, from the sea surface temperature (SST) satellite images. Our approach is based on the evalua- tion and comparison between two unsupervised classification methods, Otsu and Fuzzy C-means, and explores the appli- cability of these methods to our classification problem. The latter consists in coarse detection of the main thermal front that separates coastal cold upwelling waters from the remain- ing ocean waters. The algorithm has been applied and val- idated by an oceanographer over a database of 66 SST im- ages corresponding to southern Moroccan coastal upwelling of the years 2004, 2005, 2007 and 2009. The results indicate that the proposed algorithm revealed is promising and reli- able on different upwelling scenarios and for a wide variety of oceanographic conditions.