Machine learning approach for ship detection using remotely sensed images

Events in the past have suggested that the coastal security has to be improved and constant watch over the sea is required. Remotely sensed images being a rich source of information can be used for the same. However, the processing of remotely sensed images in order to extract the required information is a challenging task. Furthermore, the system has to be trained in order to automate the process of ship detection from the acquired images. This Paper aims onto reviewing the various existing methods for ship detection stating their advantages and limitations. It also states the experimental results obtained by using Haar-like algorithm which has been widely used in the field of image recognition. The drawbacks of this technique such as its exponential time consumption and negligence of ships in the port have been rectified with a novel methodology which uses Tensor Flow technology and Decision Boundary Feature Extraction(DBFE).