Ship Detection and Recognition Using Hybrid Combination Algorithm

摘要

Ship detection in the satellite images is gaining the attention of researchers as it has wide range of applications such as fishery management, maritime management, traffic monitoring etc. Considering the highly variable environment and weather conditions in closely spaced areas, obtaining a very controlled false alarm rate is the major problem for a non-homogenous sea clutter. since the SAR images are less influenced by weather conditions and time they are very suitable for ship detection. In a short period, the modern SARs can generate large amounts of data which brings the need for automatic detection of targets. These target detection systems carry out the detection process in three stages: identification, discrimination and classification. In this paper, we are proposing a new method of object detection by using the shape and color analysis, followed by morphological operation and GLCM. The proposed algorithm can be used to detect the crashed aero planes, floating containers and many other objects. The common CFAR has one false alarm while the results from the Hybrid combination algorithm matches with the ground truth.