Method for recognizing boats in a port includes: establishing land/water boundaries from an image of the port including one or more boats, using a land mask and image statistics, without requiring a site model; utilizing the established land/water boundaries to determine thresholds and normalization parameters; detecting one or more bow candidates for the boats in the image; computing an orientation of each of the boats with detected bows; computing features along the bow candidates to eliminate false bow candidates; fitting boat models to the detected one or more bow candidates; determining one or more best fit models from the fitted boat models; calculating features in an interior, exterior and outline of the best fit models; applying a statistical classifier to score each of the one or more best fit models and to eliminate false alarms; and eliminating lower scoring best fit models.