Application of Self-Attention Generative Adversarial Network for Electromagnetic Imaging in Half-Space

doi:10.3390/s24072322

機構典藏 > College of Engineering > Graduate Institute & Department of Electrical Engineering > Journal Article > Item 987654321/125884

Please use this identifier to cite or link to this item: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/125884

Title:	Application of Self-Attention Generative Adversarial Network for Electromagnetic Imaging in Half-Space
Authors:	Chiu, Chien-Ching;Lee, Yang-Han;Chen, Po-Hsiang;Shih, Ying-Chen;Hao, Jiang
Keywords:	inverse scattering problem;self-attention;generative adversarial network;real-time imaging;back-propagation scheme
Date:	2024-04-05
Issue Date:	2024-08-07 12:06:58 (UTC+8)
Publisher:	MDPI
Abstract:	In this paper, we introduce a novel artificial intelligence technique with an attention mechanism for half-space electromagnetic imaging. A dielectric object in half-space is illuminated by TM (transverse magnetic) waves. Since measurements can only be made in the upper space, the measurement angle will be limited. As a result, we apply a back-propagation scheme (BPS) to generate an initial guessed image from the measured scattered fields for scatterer buried in the lower half-space. This process can effectively reduce the high nonlinearity of the inverse scattering problem. We further input the guessed images into the generative adversarial network (GAN) and the self-attention generative adversarial network (SAGAN), respectively, to compare the reconstruction performance. Numerical results prove that both SAGAN and GAN can reconstruct dielectric objects and the MNIST dataset under same measurement conditions. Our analysis also reveals that SAGAN is able to reconstruct electromagnetic images more accurately and efficiently than GAN.
Relation:	Sensors 24(7), 2322
DOI:	10.3390/s24072322
Appears in Collections:	[Graduate Institute & Department of Electrical Engineering] Journal Article

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	51	View/Open

Loading...