Decision Tree Based Sea-Surface Weak Target Detection With False Alarm Rate Controllable

Aiming at accurate weak sea-surface target detection, this letter devotes to designing a learning-based detector that can work well even in varying detection environments. We first exploit the concept of the fractal theory to extract three representative features in the time and frequency domains and construct a three-dimensional feature space. We then combine the constructed feature space with the decision tree approach to design an environment-adaptive detector. Most importantly, we modify the decision tree based detector to an FAR-controllable detector to meet the requirements of different detection applications. Experimental results demonstrate that, compared with existing detectors, our proposed detector improves the detection probability in both low FAR (up to 35%) and low SCR cases (up to 55%).