Automatic Inshore Ship Detection in Satellite Imageries Using DWT and SIFT Features

Abstract

Abstract-Automatic inshore ship detection technique from satellite images is most commonly used for the purpose of military application, maritime management, and harbor traffic management, etc. Inshore ship detection from satellite images is a very useful but challenging task it is difficult to detect because of different shapes and different directions of the ship, illumination, and weather, shadow of disturbance and complex backgrounds. This paper contains an automatic inshore ship detection method that uses various technologies such as SIFT (Scale Invariant Feature Transform) and preprocessing algorithm with a discrete wavelet transform (DWT) is proposed. SIFT converts image content into local features. It also coordinates with content that is invariant to translation; rotation scale and another image parameter. SIFT features helps in matching large databases of the object to individual features and also generate different features for the small objects in local. The preprocessing algorithm with discrete wavelet transform (DWT) helps to remove the irregularities and noise to helps enhance the quality of images. The Euclidean distance helps in measuring the minimum distance between the target image and the tested image. Finally, the result of Scale Invariant Feature Transform (SIFT) and proposed method result and detect the exact detection of the inshore ship. Using our proposed technique ship detection tasks perform very precisely.Keywords-Scale Invariant Feature Transform (SIFT), Euclidean distance, Discrete wavelet