Elongated Object Orientation Estimation Based on Deep Neural Networks
- Resource Type
- Conference
- Authors
- Sun, Hai-Han; Lee, Yee Hui; Yucel, Abdulkadir C.; Ow, Genevieve; Yusof, Mohamed Lokman Mohd
- Source
- 2021 IEEE International Symposium on Antennas and Propagation and USNC-URSI Radio Science Meeting (APS/URSI) Antennas and Propagation and USNC-URSI Radio Science Meeting (APS/URSI), 2021 IEEE International Symposium on. :1998-1999 Dec, 2021
- Subject
- Fields, Waves and Electromagnetics
Deep learning
Estimation error
Ground penetrating radar
Spaceborne radar
Neural networks
Meetings
Radar imaging
ground-penetrating radar
multi-polarization
multi-mask neural network
orientation estimation
elongated subsurface object
- Language
The horizontal and vertical orientation angles of an elongated subsurface object are key parameters for object identification and imaging in ground-penetrating radar (GPR) applications. Conventional methods can only extract the horizontal orientation angle or estimate both angles in narrow ranges. To address these issues, we present a multi-polarization aggregation and selection neural network (MASNet) to estimate both angles of an elongated subsurface object in the entire spatial range. The network takes the multi-polarimetric radargrams as inputs, integrates their characteristics in the feature space, and selects discriminative features of reflected signal patterns for accurate orientation estimation. Numerical results show that the MASNet achieves high estimation accuracy with an angle estimation error of less than 5°. The promising results obtained in the study encourages new solutions for GPR-related tasks by integrating multi-polarization information with deep learning techniques.