Extraction of object features from high resolution SAR images based on SURF features

Because of the influence of object orientation, sensor parameters and environmental factors, objects in SAR images have high changeability when the images were formed. The complex environment makes it difficult to extract object features we aim at. In this paper we take aircraft for example, propose a method to extract object features in high resolution SAR images. The features involve SURF features, numbers of engines, and the angle between engines and principal axes. Finally using the features combination we prove that the features extracted can effectively describe objects in various conditions.