Extraction And Detection Of Target Trajectory In Bearing-time Records Under Complex Condition

During the sonar signal processing procedure,the bearing-time records(BTRs map)is relatively wide used in sonar signal recording and display.With advances in technology in Noise reduction,The noise of modern ships are gradually reduce which make the SNR(Signal to noise ratio)of the BTRs map become smaller and smaller,and this brings a great challenge to the target trajectory extraction.This paper mainly aims at the extraction and detection of the target trajectory in Bearing-Time Records under complex conditions.To address the problem that the weak target trajectory is difficult to extract under strong background noise in bearing-time records and it is hard to extract the target.According to the characteristics of the target trajectory,this paper proposed a ridge signal enhancement algorithm.Firstly,this algorithm uses partial localized filtering to suppress some of the gentle background noise,then,Enhance the target trajectory by cross-correlation filter to improve the signal-to-noise ratio of the image,and finally,Integrate in a small area to achieve the purpose of connecting the broken part in target trajectory.The experimental results show that the algorithm has a very good enhancement effect on the target trajectory in bearing-time records,and the results of experiments also show that this method is superior to OTA algorithm and MMF algorithm.In bearing-time records,the target trajectory may appear to broken,overlapping and so on,combined with the geometric characteristics of the target trajectory motion information,this paper proposes a trajectory correlation and trajectory correction method for suspected target points to suppressing background noise and completing the target track.The simulation experiment results show that this method can effectively deal with the interference of background noise and accurately connect the target trajectory.For the emergence of false target trajectory in extraction results,this paper proposes three Classification feature dimensions by analyzing the differences between the real target trajectory and the false target trajectory in the gray scale and geometric information.Those classification feature dimensions are the average of peak to neighbor of the trajectory,the variance of peak to neighbor of the trajectory,and the trajectory completion.To achieve the effect of suppressing the false target trajectory,this paper uses the support vector machine to classifying the extraction results,and the experiment results show that it can effectively classify the false target trajectory,the experiment also compare it to the spectral clustering algorithm.And it shows that the support vector machine has a good effect on the suppression of the false target trajectory.