Deep-learning Linear Sampling Method for Shape Restoration of Multilayered Scatterers

Kuo, Yu-Hsin; Kiang, Jean‐Fu

doi:10.2528/pierc22081005

Cited by 2 publications

(4 citation statements)

References 21 publications

(44 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our aim for displaying the isosurfaces in this manner is to allow straightforward visual evaluation of the differences in reconstruction across the four methods under investigation. Selecting the best image thresholds when the true target geometry is unknown is an interesting problem [33] that is outside the scope of this study. The isosurfaces generated for the G-shaped target are plotted in Fig.…”

Section: Visualization Of 3d Resultsmentioning

confidence: 99%

“…The linear sampling method (LSM) is a technique for reconstructing target shape that has received significant attention in the literature due to several potential advantages (e.g., [20][21][22][23][24][25][26][27][28][29][30][31][32][33]). The LSM involves solving for a set of transmit weights that focus equivalent currents in the domain [24] as opposed to directly solving for the target shape.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Linear Sampling Method Imaging of Three-dimensional Conducting Targets from Limited Apertures via Phase-delay-constrained Formulations

Burfeindt,

Alqadah

2023

PIER

View full text Add to dashboard Cite

The linear sampling method (LSM) is a qualitative inverse scattering technique for reconstructing the shape of a target. It has several beneficial qualities, including the avoidance of nonlinear optimization and simplified scattering approximations. However, it often struggles when sensors can only be placed on one side of the target. In this paper, we investigate two alternative LSM formulations for overcoming the limited-aspect challenge. The first, the phase-delay frequency variation LSM (PDFV-LSM), incorporates coherent processing across frequency to improve discrimination in the range direction. The second, the phase-encoded LSM (PE-LSM), enhances the PDFV-LSM approach with a receive-beamforming operation to decrease the complexity of the inverse problem. We apply both techniques to simulated data from three-dimensional targets and three-dimensional limited-aspect arrays. We generate three-dimensional reconstructions and compare them to reconstructions from both the conventional LSM and conventional backprojection-based processing. The results demonstrate superior reconstruction fidelity for either the PDFV-LSM, PE-LSM, or both, across a wide variety of imaging scenarios due to finer range resolution. They also demonstrate trade-offs between the two enhanced LSM techniques, with the PE-LSM achieving better range resolution and robustness to noise and the PDFV-LSM achieving better lateral resolution.

show abstract

Section: Visualization Of 3d Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Linear Sampling Method Imaging of Three-dimensional Conducting Targets from Limited Apertures via Phase-delay-constrained Formulations

Burfeindt,

Alqadah

2023

PIER

View full text Add to dashboard Cite

show abstract

“…Better reconstruction is aided by the merging of low-level and high-level fea- tures made possible by the skip connections between relevant encoder and decoder levels. More details about the U-Net architecture used in this work can be found in [17], [29].…”

Section: Deep Learning Assisted Linear Sampling Methods For Inverse P...mentioning

confidence: 99%

“…In [16], the orthogonality sampling method (OSM) is used along with deep learning architecture called the U-Net to reconstruct the permittivites of the dielectric objects. In [17], a deep learning-assisted linear sampling method (DLSM) is proposed for the reconstruction of multi-layered dielectric objects with cylindrical and rectangular cross-sections. So far, this is the only work reported on LSM with deep learning.…”

Section: Introductionmentioning

confidence: 99%

Reconstruction of Mixed Boundary Objects and Classification Using Deep Learning and Linear Sampling Method

Harisha,

Mallikarjun,

Amit

2024

RADIOENGINEERING

View full text Add to dashboard Cite

The linear sampling method is a simple and reliable linear inversion technique for determining the morphological features of unknown objects under investigation. Nevertheless, there are many challenges that this method depends on the frequency of operation and it is unable to produce satisfactory results for objects with complex shapes. This paper proposes a hybrid model, which combines conventional linear sampling method and deep learning for the reconstruction of mixed boundary objects. In this approach, the initial approximation of mixed boundary objects derived from linear sampling method serves as the training data for the U-Net based convolutional neural network. The network then learns to correlate this approximation with the corresponding ground truth profiles. Along with the reconstruction of mixed boundary objects, they are also classified as dielectric or conductor, and count of each object type are measured. Furthermore, the low-frequency and highfrequency characteristics of the linear sampling method are analyzed, and its limitations are overcome by combining it with a deep learning approach. The effectiveness of the proposed model is validated using several examples of synthetic and experimental data. The results demonstrate that the proposed method outperforms the conventional Linear sampling method in terms of accuracy.

show abstract

Deep-learning Linear Sampling Method for Shape Restoration of Multilayered Scatterers

Cited by 2 publications

References 21 publications

Linear Sampling Method Imaging of Three-dimensional Conducting Targets from Limited Apertures via Phase-delay-constrained Formulations

Linear Sampling Method Imaging of Three-dimensional Conducting Targets from Limited Apertures via Phase-delay-constrained Formulations

Reconstruction of Mixed Boundary Objects and Classification Using Deep Learning and Linear Sampling Method

Contact Info

Product

Resources

About