Multi-target track initiation method based on support vector machine in non-uniform strong clutter environment

Abstract

Aiming at the problem that the track initiation performance is easily affected by the complex background information in the non-uniform and strong environment, this paper proposes an improved multi-target track initiation method based on support vector machine. First, we use the intuitive method for rough start. Second, we adopt support vector machine to improve the results by filtering out false tracks. And the intuitive method's multiple track initial results are used as training data to train the support vector machine. Third, for the non-uniform clutter environment, the models are separately trained and saved for different clutter density environments in advance, and the detection area is partitioned according to the clutter density at the beginning of the track. Finally, different models are used in different regions. The simulation results show that compared with the traditional track initiation method and the support vector machine method using a single model, this algorithm improves the correct track initiation rate, reduces the false track occupancy rate, and effectively improves the algorithm performance of multi-target aircraft initiation in environment with non-uniform and strong clutter.