Electromagnetic imaging of a dielectric micro-structure via convolutional neural networks. Ran, P., Qin, Y., & Lesselier, D. In 2019 27th European Signal Processing Conference (EUSIPCO), pages 1-5, Sep., 2019. Paper doi abstract bibtex Convolutional neural networks (CNN) are applied to the time-harmonic electromagnetic diagnostic of a dielectric micro-structure. The latter consists of a finite number of circular cylinders (rods) with a fraction of wavelength radius that are set parallel to and at sub-wavelength distance from one another. Discrete scattered fields are made available around it in a free-space multisource-multireceiver configuration. The aim is to characterize this micro-structure, like positions of rods or their absence, and in effect to map their dielectric contrasts w.r.t. the embedding space. A computationally efficient field representation based on a method of moments (MoM) is available to model the field. Iterative, sparsity-constrained solutions work well to find missing rods, but may lack generality and need strong priors. As for time-reversal and like noniterative solutions, they may fail to capture the scattering complexity. These limitations can be alleviated by relying on deep learning concepts, here via convolutional neural networks. How to construct the inverse solver is focused onto. Representative numerical tests illustrate the performance of the approach in typical situations. Comparisons with results from a contrast-source inversion (CSI) introduced in parallel are performed. Emphasis is on potential super-resolution in harmony with subwavelength features of the micro-structure.
@InProceedings{8903073,
author = {P. Ran and Y. Qin and D. Lesselier},
booktitle = {2019 27th European Signal Processing Conference (EUSIPCO)},
title = {Electromagnetic imaging of a dielectric micro-structure via convolutional neural networks},
year = {2019},
pages = {1-5},
abstract = {Convolutional neural networks (CNN) are applied to the time-harmonic electromagnetic diagnostic of a dielectric micro-structure. The latter consists of a finite number of circular cylinders (rods) with a fraction of wavelength radius that are set parallel to and at sub-wavelength distance from one another. Discrete scattered fields are made available around it in a free-space multisource-multireceiver configuration. The aim is to characterize this micro-structure, like positions of rods or their absence, and in effect to map their dielectric contrasts w.r.t. the embedding space. A computationally efficient field representation based on a method of moments (MoM) is available to model the field. Iterative, sparsity-constrained solutions work well to find missing rods, but may lack generality and need strong priors. As for time-reversal and like noniterative solutions, they may fail to capture the scattering complexity. These limitations can be alleviated by relying on deep learning concepts, here via convolutional neural networks. How to construct the inverse solver is focused onto. Representative numerical tests illustrate the performance of the approach in typical situations. Comparisons with results from a contrast-source inversion (CSI) introduced in parallel are performed. Emphasis is on potential super-resolution in harmony with subwavelength features of the micro-structure.},
keywords = {convolutional neural nets;electromagnetic wave scattering;inverse problems;iterative methods;learning (artificial intelligence);method of moments;electromagnetic imaging;dielectric microstructure;convolutional neural networks;time-harmonic;wavelength radius;subwavelength distance;discrete scattered fields;free-space multisource-multireceiver configuration;dielectric contrasts;computationally efficient field representation;missing rods;CNN;Permittivity;Inverse problems;Nickel;Dielectrics;Mathematical model;Europe;Signal processing;convolutional neural networks;micro-structure;super-resolution imaging;inverse scattering problem},
doi = {10.23919/EUSIPCO.2019.8903073},
issn = {2076-1465},
month = {Sep.},
url = {https://www.eurasip.org/proceedings/eusipco/eusipco2019/proceedings/papers/1570531220.pdf},
}
Downloads: 0
{"_id":"FrGvSADx3BXEuWmFG","bibbaseid":"ran-qin-lesselier-electromagneticimagingofadielectricmicrostructureviaconvolutionalneuralnetworks-2019","authorIDs":[],"author_short":["Ran, P.","Qin, Y.","Lesselier, D."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","author":[{"firstnames":["P."],"propositions":[],"lastnames":["Ran"],"suffixes":[]},{"firstnames":["Y."],"propositions":[],"lastnames":["Qin"],"suffixes":[]},{"firstnames":["D."],"propositions":[],"lastnames":["Lesselier"],"suffixes":[]}],"booktitle":"2019 27th European Signal Processing Conference (EUSIPCO)","title":"Electromagnetic imaging of a dielectric micro-structure via convolutional neural networks","year":"2019","pages":"1-5","abstract":"Convolutional neural networks (CNN) are applied to the time-harmonic electromagnetic diagnostic of a dielectric micro-structure. The latter consists of a finite number of circular cylinders (rods) with a fraction of wavelength radius that are set parallel to and at sub-wavelength distance from one another. Discrete scattered fields are made available around it in a free-space multisource-multireceiver configuration. The aim is to characterize this micro-structure, like positions of rods or their absence, and in effect to map their dielectric contrasts w.r.t. the embedding space. A computationally efficient field representation based on a method of moments (MoM) is available to model the field. Iterative, sparsity-constrained solutions work well to find missing rods, but may lack generality and need strong priors. As for time-reversal and like noniterative solutions, they may fail to capture the scattering complexity. These limitations can be alleviated by relying on deep learning concepts, here via convolutional neural networks. How to construct the inverse solver is focused onto. Representative numerical tests illustrate the performance of the approach in typical situations. Comparisons with results from a contrast-source inversion (CSI) introduced in parallel are performed. Emphasis is on potential super-resolution in harmony with subwavelength features of the micro-structure.","keywords":"convolutional neural nets;electromagnetic wave scattering;inverse problems;iterative methods;learning (artificial intelligence);method of moments;electromagnetic imaging;dielectric microstructure;convolutional neural networks;time-harmonic;wavelength radius;subwavelength distance;discrete scattered fields;free-space multisource-multireceiver configuration;dielectric contrasts;computationally efficient field representation;missing rods;CNN;Permittivity;Inverse problems;Nickel;Dielectrics;Mathematical model;Europe;Signal processing;convolutional neural networks;micro-structure;super-resolution imaging;inverse scattering problem","doi":"10.23919/EUSIPCO.2019.8903073","issn":"2076-1465","month":"Sep.","url":"https://www.eurasip.org/proceedings/eusipco/eusipco2019/proceedings/papers/1570531220.pdf","bibtex":"@InProceedings{8903073,\n author = {P. Ran and Y. Qin and D. Lesselier},\n booktitle = {2019 27th European Signal Processing Conference (EUSIPCO)},\n title = {Electromagnetic imaging of a dielectric micro-structure via convolutional neural networks},\n year = {2019},\n pages = {1-5},\n abstract = {Convolutional neural networks (CNN) are applied to the time-harmonic electromagnetic diagnostic of a dielectric micro-structure. The latter consists of a finite number of circular cylinders (rods) with a fraction of wavelength radius that are set parallel to and at sub-wavelength distance from one another. Discrete scattered fields are made available around it in a free-space multisource-multireceiver configuration. The aim is to characterize this micro-structure, like positions of rods or their absence, and in effect to map their dielectric contrasts w.r.t. the embedding space. A computationally efficient field representation based on a method of moments (MoM) is available to model the field. Iterative, sparsity-constrained solutions work well to find missing rods, but may lack generality and need strong priors. As for time-reversal and like noniterative solutions, they may fail to capture the scattering complexity. These limitations can be alleviated by relying on deep learning concepts, here via convolutional neural networks. How to construct the inverse solver is focused onto. Representative numerical tests illustrate the performance of the approach in typical situations. Comparisons with results from a contrast-source inversion (CSI) introduced in parallel are performed. Emphasis is on potential super-resolution in harmony with subwavelength features of the micro-structure.},\n keywords = {convolutional neural nets;electromagnetic wave scattering;inverse problems;iterative methods;learning (artificial intelligence);method of moments;electromagnetic imaging;dielectric microstructure;convolutional neural networks;time-harmonic;wavelength radius;subwavelength distance;discrete scattered fields;free-space multisource-multireceiver configuration;dielectric contrasts;computationally efficient field representation;missing rods;CNN;Permittivity;Inverse problems;Nickel;Dielectrics;Mathematical model;Europe;Signal processing;convolutional neural networks;micro-structure;super-resolution imaging;inverse scattering problem},\n doi = {10.23919/EUSIPCO.2019.8903073},\n issn = {2076-1465},\n month = {Sep.},\n url = {https://www.eurasip.org/proceedings/eusipco/eusipco2019/proceedings/papers/1570531220.pdf},\n}\n\n","author_short":["Ran, P.","Qin, Y.","Lesselier, D."],"key":"8903073","id":"8903073","bibbaseid":"ran-qin-lesselier-electromagneticimagingofadielectricmicrostructureviaconvolutionalneuralnetworks-2019","role":"author","urls":{"Paper":"https://www.eurasip.org/proceedings/eusipco/eusipco2019/proceedings/papers/1570531220.pdf"},"keyword":["convolutional neural nets;electromagnetic wave scattering;inverse problems;iterative methods;learning (artificial intelligence);method of moments;electromagnetic imaging;dielectric microstructure;convolutional neural networks;time-harmonic;wavelength radius;subwavelength distance;discrete scattered fields;free-space multisource-multireceiver configuration;dielectric contrasts;computationally efficient field representation;missing rods;CNN;Permittivity;Inverse problems;Nickel;Dielectrics;Mathematical model;Europe;Signal processing;convolutional neural networks;micro-structure;super-resolution imaging;inverse scattering problem"],"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"https://raw.githubusercontent.com/Roznn/EUSIPCO/main/eusipco2019url.bib","creationDate":"2021-02-11T19:15:22.122Z","downloads":0,"keywords":["convolutional neural nets;electromagnetic wave scattering;inverse problems;iterative methods;learning (artificial intelligence);method of moments;electromagnetic imaging;dielectric microstructure;convolutional neural networks;time-harmonic;wavelength radius;subwavelength distance;discrete scattered fields;free-space multisource-multireceiver configuration;dielectric contrasts;computationally efficient field representation;missing rods;cnn;permittivity;inverse problems;nickel;dielectrics;mathematical model;europe;signal processing;convolutional neural networks;micro-structure;super-resolution imaging;inverse scattering problem"],"search_terms":["electromagnetic","imaging","dielectric","micro","structure","via","convolutional","neural","networks","ran","qin","lesselier"],"title":"Electromagnetic imaging of a dielectric micro-structure via convolutional neural networks","year":2019,"dataSources":["NqWTiMfRR56v86wRs","r6oz3cMyC99QfiuHW"]}