Semi-automated maltese front position determination

We investigate the reliability of a semi-automated system designed to locate the Maltese front from satellite AVHRR and SAR imagery. The opening and closing operations of mathematical morphology afford a means of image segmentation that provides smooth, strong, continuous edge except where the edges are obscured by clouds or similar phenomena. We evaluate the results by comparing them with front positions found by experienced Navy analysts. Our system provides an objective method for front location that is less labor-intensive than manual methods currently in use.