Research On Air Moving Target Confirmation Method Based On Space-Time Trajectory Template Matching

With the rapid development of aerospace technology,the flight speed of air targets has increased and the stealth performance has been enhanced,making it more difficult to confirm targets.Strengthening air control and developing air defense forces are important means for protecting national space resources and safeguarding the security of national airspace,and are also important prerequisites for ensuring the stable development of the national economy and long-term security of the country.It is of great significance to build an air target supervision system and an air defense early warning system.Air target confirmation is an important part of air monitoring and early warning system.At the same time,with the improvement of mobile computing capabilities and the development of position acquisition technology,the detection system can acquire high-frequency and high-precision target space-time trajectory data.The spatio-temporal trajectory data sequence not only records the target's spatial position information,but also reflects the target's movement law.It can be used for aerial target confirmation in combination with the target's prior information.However,the traditional aerial target confirmation method based on spatiotemporal trajectory template matching only treats the trajectory as a simple time series,which is limited to the similarity measurement of the spatial distance,does not fully exploit the temporal characteristics of the trajectory data,and does not have high-precision applications.Level.The space-time trajectory data of the target under space-based remote sensing also contains the radiation information of the target,but few people currently integrate the space-time radiation information to confirm the target.This article is aimed at the above problems,by fully analyzing the space,time and radiation characteristics of the target and the correlation between them,to study the aerial target confirmation method based on the template matching of space-time trajectory and carry out experimental verification,the main research work is as follows:(1)Research on modeling method of air target movement trajectory.Due to the long detection distance of the space-based remote sensing detection platform and the low spatial resolution of the detector,aerial targets usually appear as point targets or spot targets with a size of several pixels on the image,and do not have geometric morphological characteristics and surface texture information.In this paper,the aerial target motion trajectory simulation model is established to measure and characterize the aerial target under different detection conditions,and the detection platform vibration model is established to analyze the impact of the detection platform jitter on the trajectory data.Finally,simulation experiments are carried out based on the existing target parameters.Based on the simulation data,the differences in the movement characteristics of different types of air targets are analyzed,which provides a theoretical basis for the study of target confirmation methods.(2)Research on similarity measurement method based on normalized trajectory template.According to the space-time motion characteristics of aerial targets,the methods of aerial target confirmation are studied.Firstly,the principle comparison analysis of the classical trajectory similarity measurement method is carried out,and the different methods are tested through the simulation data established by the air target movement trajectory model.The experimental results show that the traditional method has low target confirmation accuracy.In view of the above problems,this paper proposes an improved Hausdorff distance target confirmation method.By setting the motion direction angle threshold to segment the trajectory again,the Hausdorff distance method is susceptible to noise interference and experiments are carried out.The experimental results show that the similarity measurement method proposed in this paper has high computational efficiency and effectively improves the accuracy of target confirmation.(3)Research on the method of HMM-PNN target confirmation integrating the radiation characteristics.Considering that improving the Hausdorff distance is still only concerned with the problem of spatial position information,there is no problem of the difference in motion characteristics and radiation characteristics between different air targets in the time dimension.This paper is based on the infrared detection mechanism for the radiation characteristics of different types of air targets.Analyze the cognition,start from the correlation between the spatiotemporal radiation characteristics,take advantage of the low data requirements of the hidden Markov model,establish a hidden Markov model of speed,acceleration,position and radiation,and learn the spatiotemporal trajectory data And processing,using the probabilistic neural network to judge the target class of the maximum likelihood trajectory obtained by the hidden Markov model,to complete the target confirmation.This method inherits the advantages of the hidden Markov model in processing time series data,and combines the strong classification and recognition characteristics of the probabilistic neural network,which can better deal with the correlation between features of different dimensions.(4)Experimental verification and performance evaluation.The space-time trajectory data of the air target is used to verify the HMM-PNN target confirmation method that combines the space-time and radiation characteristics,and the statistics obtain the recall and precision.The experimental results show that the method has achieved a precision of 93.67% and 94.80.The recall rate of% is greatly improved compared with the target confirmation method of improving the Hausdorff distance;the evaluation index of the current target confirmation method cannot differentially evaluate the confirmation results of different information trajectories.This paper introduces the modulation factor Give different weights to the trajectories of different information,and then achieve the improvement of the performance evaluation index of recall rate and precision rate,and conduct experimental verification.The experimental results show that the evaluation index can reasonably evaluate the impact of data accumulation on target confirmation.