Open AccessApplication of feature modelling method based on SAR image in target interpretation
Author(s) -
Li Caiping,
Zhou Xiaoming,
Chang Lei
Publication year - 2019
Publication title -
the journal of engineering
Language(s) - English
Resource type - Journals
ISSN - 2051-3305
DOI - 10.1049/joe.2019.0206
Subject(s) - computer science , artificial intelligence , feature (linguistics) , projection (relational algebra) , computer vision , pattern recognition (psychology) , image (mathematics) , inverse , back projection , graph , algorithm , mathematics , philosophy , linguistics , geometry , theoretical computer science
This study starts with the engineering modelling of SAR image features, analyses the platform elements that affect the image features, and designs the acquisition flow of samples and parameters. The two‐dimensional SAR image of the target is processed by three‐dimensional inverse projection and projected to the geometric model of the candidate target, and the three‐dimensional inverse projection scattering graph of the target is obtained. At the same time, the high‐frequency electromagnetic scattering calculation method is used to predict the 3D hot spot distribution of each candidate target, to improve the recognition accuracy of the target, and to achieve the goal of target interpretation. The actual results show that the target interpretation method based on SAR imaging feature modelling proposed here can effectively improve the target recognition rate.