EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Learned-SVD Approach to the Electromagnetic Inverse Source Problem

Amedeo Capozzoli, Ilaria Catapano iD, Eliana Cinotti iD, Claudio Curcio iD, Giuseppe Esposito iD, Gianluca Gennarelli, Angelo Liseno, Giovanni Ludeno iD, Francesco Soldovieri iD

Sensors · 11.07.2024 · 24 (14) · 4496

DOI: 10.3390/s24144496 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 11.07.2024. Verfasst von Amedeo Capozzoli und 8 weiteren Autor:innen. Beteiligte Einrichtung: Dipartimento di Ingegneria Elettrica e delle Tecnologie dell'Informazione (DIETI), Università di Napoli Federico II, Via Claudio 21, I 80125 Napoli, …. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 63 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: This work has been supported by the project "Modeling and deep learning in inverse scattering problems: a new paradigm" funded by the MIUR Progetti di Ricerca…, MIUR Progetti di Ricerca di Rilevante Interesse Nazionale (PRIN) Bando 2022.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We propose an artificial intelligence approach based on deep neural networks to tackle a canonical 2D scalar inverse source problem. The learned singular value decomposition (L-SVD) based on hybrid autoencoding is considered. We compare the reconstruction performance of L-SVD to the Truncated SVD (TSVD) regularized inversion, which is a canonical regularization scheme, to solve an ill-posed linear inverse problem. Numerical tests referring to far-field acquisitions show that L-SVD provides, with proper training on a well-organized dataset, superior performance in terms of reconstruction errors as compared to TSVD, allowing for the retrieval of faster spatial variations of the source. Indeed, L-SVD accommodates a priori information on the set of relevant unknown current distributions. Different from TSVD, which performs linear processing on a linear problem, L-SVD operates non-linearly on the data. A numerical analysis also underlines how the performance of the L-SVD degrades when the unknown source does not match the training dataset.

Schlagwörter

Stichwörter der Autor:innen

autoencoderdeep neural networksinverse sourcelearned singular value decompositionsingular value decomposition

Autor:innen und Institutionen

  • Dipartimento di Ingegneria Elettrica e delle Tecnologie dell'Informazione (DIETI), Università di Napoli Federico II, Via Claudio 21, I 80125 Napoli, Italy.
  • Consiglio Nazionale delle Ricerche, Istituto per il Rilevamento Elettromagnetico dell'Ambiente (IREA), Via Diocleziano 328, I 80124 Napoli, Italy.

ORCID-Profile: Ilaria Catapano, Eliana Cinotti, Claudio Curcio, Giuseppe Esposito, Giovanni Ludeno, Francesco Soldovieri

Förderung

  • This work has been supported by the project "Modeling and deep learning in inverse scattering problems: a new paradigm" funded by the MIUR Progetti di Ricerca…
  • MIUR Progetti di Ricerca di Rilevante Interesse Nazionale (PRIN) Bando 2022

Zitieren

APA

Capozzoli, A., Catapano, I., Cinotti, E., Curcio, C., Esposito, G., Gennarelli, G., Liseno, A., Ludeno, G., & Soldovieri, F. (2024). A Learned-SVD Approach to the Electromagnetic Inverse Source Problem. Sensors, 24(14), 4496. https://doi.org/10.3390/s24144496

Vancouver

Capozzoli A, Catapano I, Cinotti E, Curcio C, Esposito G, Gennarelli G, et al. A Learned-SVD Approach to the Electromagnetic Inverse Source Problem. Sensors. 2024;24(14):4496. doi:10.3390/s24144496

BibTeX
@article{Capozzoli2024,
  author = {Capozzoli, Amedeo and Catapano, Ilaria and Cinotti, Eliana and Curcio, Claudio and Esposito, Giuseppe and Gennarelli, Gianluca and Liseno, Angelo and Ludeno, Giovanni and Soldovieri, Francesco},
  title = {A Learned-SVD Approach to the Electromagnetic Inverse Source Problem},
  journal = {Sensors},
  year = {2024},
  volume = {24},
  number = {14},
  pages = {4496},
  doi = {10.3390/s24144496}
}
Kostenlos starten →