A low altitude small target detection method based on Feature Clutter Map

Abstract

Sea clutter changes dynamically with the external environment, and the detection of small low-altitude targets is seriously affected by sea clutter. The existing sea clutter and target feature classification methods based on machine learning have the problems of limited feature dimensions and the training model is seriously affected by uneven samples. This paper proposed a method for detecting low altitude small targets on the sea surface based on the feature clutter map. By extracting the multi-dimensional features of the target and sea clutter, this method establishes the feature clutter map under the pure sea clutter background, and determines whether there is a target by comparing with the feature clutter map during target detection. Compared with the target decision method based on three-dimensional convex hull, the feature dimension of this method can be extended to avoid missing detection and false alarm caused by different sensitivity of data to different features. Moreover, the pure sea clutter samples are easy to obtain, and there is no problem of inaccurate judgment caused by the imbalance between the target and the sea clutter training samples, so as to improve the detection probability of low altitude small targets under the sea clutter background. The method is verified by the measured data of a ku-band radar, and can effectively detect a low altitude small target in the background of sea clutter.