Multi-evidence fusion recognition of ship targets in sea battlefield’s remote sensing images

Abstract

Abstr act. Reliably recognizing ship targets in the sea battlefield has become an increasingly pressing need, as the capabilities for image acquisition are growing rapidly. In the research, a modified ship targets fusion recognition model based on the Dempster-Shafer evidence theory is proposed. The algorithm firstly detects ship targets from the sea battlefield's remote sensing image. Then extracts the multiple image features of these target candidate areas as the evidence to recognize the ship targets. Finally, recognizes the ship targets from the image using the Dempster-Shafer evidence theory based on multiple ship features, and sends the recognition result. Experiment show that this method can be used to reliably and effectively recognize targets information in the sea battlefield. Keywor ds: target recognition; Dempster-Shafer evidence; ship features; remote sensing image 1. Instr uction As one of the main areas of operations of modern war, Sea battle field situation vary from minute to minute in battle. Ship targets are the key targets of the maritime monitoring and wartime attack. Rapid and accurate identification of ship sea battlefield tactics and provide support for decision-making for commanders, which is greatly related to the success or failure of the battle. Many judging methods of warship target intention in a sea battle, they are built on the basis of the target recognition. In modern sea battles, in order to make the situation assessment and battlefield decision, we should first solve how to recognize accurately different kinds of battle ships in the distance.