Esearch on UAV location and tracking technology based on radar detection
To improve the recognition accuracy of drones,a DSAE-VGG drone recognition model was proposed based on Deep Sparse Auto Encoder(DSAE)and VGG networks.DSAE was used to reduce the feature dimension of the UAV data detected by radar,and the VGG network structure was designed and improved according to the UAV data detected by radar,and the improved VGG network was used to classify and identify the data after feature dimensionality reduction,so as to realize the effective identification of UAV.The simulation results showed that the proposed model could accurately identify single rotor,four rotor,and six rotor unmanned aerial vehicles,and had a high accuracy rate,with an average recognition accuracy of 97% .Compared with the CNN model and VGG model,the average recognition accuracy had been improved by 8.24% and 4.22%,respectively.It has certain advantages and can meet the accuracy requirements of radar detection based unmanned aerial vehicle recognition.
万方数据
维普
