International Journal of Emerging trends in Engineering and Development Issue 2, Vol.7 (November 2012) Available online on http://www.rspublication.com/ijeted/ijeted_index.htm ISSN 2249-6149 Page 543 Classification of ship targets using neural classifiers based on Simulated SAR images P.Subashini #1 M.Krishnaveni #2 A.Vanitha #3 Associate Professor Research Assistant Research Scholar #1,#2,#3 Department of Computer Science Avinashilingam Institute for Home Science and Higher Education for Women Coimbatore,TamilNadu,India __________________________________________________ ABSTRACT A coastal surveillance system maneuvers in a predefined area which is intended to detect a suspicious object that appears within this area. Ship detection and classification are important to maintain a recognized Maritime Picture (RMP) for surveillance systems. An automatic detection and classification system leads to some specific problems since a coastal environment is a random changing background caused by water waves and light changes. Moreover, there are spatial and temporal changes characterized by a large dynamic range of radar and the image frames captured are very expensive. By using computer-based ship SAR image simulation an extensive image database can be generated. This affords a very cost effective method for acquiring radar images when compared to the alternative methods of acquiring real images, using vessels or ship models. A classification system is modeled for vessel classification based on neural classifiers. The research work proposed in this paper describes the development of ship classification program that is designed to analyze processed simulated SAR imagery rapidly and inexpensively through image frames using image processing techniques. The simulated SAR image background are generated by radar parameters and Back projection algorithm in which image fusion technique is used to derive the simulated SAR ship images. Classification system is constructed based on two statistical classifiers PNN and KNN. The customizable framework is designed for the simulation, implementation, and verification of vessel classification using simulated SAR images. Keywords : Synthetic Aperture Radar images, radar parameters, back projection algorithm, image fusion, Statistical classifiers, simulation . ___________________________________________________________________________ Corresponding Author: P.Subashini INTRODUCTION Classification of ship targets is initially intended to help ships from collisions, as well as assisting port authorities to better control sea traffic .AIS transponders on board vessels include a GPS (Global Positioning System) receiver, which collects position, movement details and vessel structural information .Vessel classification therefore essentially comes down to vessel size estimation which shows considerably more variation in their classification performance than in their detection performance. The data collection for the