Research On Algorithm Of Marine Radar Target Detection And Tracking

With the rapid development of marine transportation industry,more and more ship sail on sea,marine transportation safety issues have attracted people's more and more attention.In order to ensure the safety of marine transportation,Vessel Traffic Services system(VTS system)is introduced into major harbors at home and abroad.The VTS system monitors ships movements of sea area by radar,and its core processing algorithm is radar target detection and tracking algorithm.In order to solve the problem that current radar target detection and tracking algorithm has weak detection ability and poor tracking ability,this thesis studies the marine radar target detection and tracking algorithm by building the platform to collect marine radar measured data.This thesis mainly includes the following three aspects:In the aspect of target detection algorithm,this thesis firstly introduces the whole process of obtaining the measured data of marine radar,and obtains the research data set of the whole thesis.Then,using the parameter estimation and hypothesis test theory to analyze the sea clutter measured data set,and obtains the optimal distribution model of sea clutter.Finally,according to the optimal distribution model of sea clutter,analyzes the principle of the corresponding CFAR detection algorithm in detail,and the measured data set is processed by this CFAR detection algorithm,and most of the sea clutter information is successfully filtered out.In the aspect of target tracking algorithm,this thesis mainly studies algorithms based on probability hypothesis density particle filter(PHD-PF).Aiming at the problem that the traditional PHD-PF algorithm is difficult to process the real data,this thesis proposes a UPDPHD-PF algorithm based on uniform particle distribution.Then,this thesis demonstrates tracking performance of the proposed al.gorithm by numerical simulation,and the simulation results show that the proposed algorithm has similar tracking performance to the traditional algorithm under the same conditions.Finally,the proposed algorithm is used to deal with the measured data set,and gets a good tracking effect.Aiming at the problem of real-time performance of the traditional PHD-PF algorithm and the UPDPHD-PF algorithm,this thesis proposes an APDPHD-PF algorithm based on adaptive particle distribution.The proposed APDPHD-PF algorithm generates new particles according to observations of the current time,so it can reduce the number of particles effectively,and reduce the computational cost and running time of the algorithm.Then,this thesis demonstrates that the proposed algorithm can improve the real-time performance while keeping the tracking performance by numerical simulation.Finally,the proposed algorithm is used to deal with the measured data set,and obtains a good tracking effect.At the same time,compared with the UPDPHD-PF algorithm,the proposed algorithm has a great advantage in real-time performance.