Synthetic aperture radar aircraft target identification method based on self-attention network
The invention belongs to the technical field of radar remote sensing application, and particularly relates to a synthetic aperture radar aircraft target identification method based on a self-attention network. The implementation process comprises the steps of obtaining a data set; the method compris...
Gespeichert in:
Hauptverfasser: | , , , |
---|---|
Format: | Patent |
Sprache: | chi ; eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The invention belongs to the technical field of radar remote sensing application, and particularly relates to a synthetic aperture radar aircraft target identification method based on a self-attention network. The implementation process comprises the steps of obtaining a data set; the method comprises the following steps of: constructing a network model based on swin transformer; inputting the training set into a network model, and extracting features by using the network model; calculating a loss function of the extracted features in a back propagation process, and adjusting network model parameters according to the loss function to obtain a final target recognition model; and inputting an SAR image needing to be detected into the final target recognition model to obtain a detection result. According to the method, classification and regression branch information is utilized to clearly decompose and jointly represent the target positioning quality, and the recognition precision can be improved while the calc |
---|