| Autor: |
Joe LoVetri, Ian Jeffrey, Liam Sparling, Nicholas Geddert |
| Rok vydání: |
2018 |
| Předmět: |
|
| Zdroj: |
2018 International Workshop on Computing, Electromagnetics, and Machine Intelligence (CEMi). |
| DOI: |
10.1109/cemi.2018.8610613 |
| Popis: |
The contrast source inversion (CSI), a robust technique for solving inverse electromagnetic scattering problems, has previously been applied to data from multiple frequencies in two primary ways: frequency-hopping solves one inverse problem at a single frequency and uses the solution as an initial guess for another frequency, while simultaneous frequency inversion considers data from multiple frequencies during a single inversion run. In this work, we present a study of the effects of frequency weighting parameters on the convergence and quality of CSI reconstruction results for microwave imaging. The study proposes schemes which attempt to exploit lower frequency data during initial iterations while including higher frequency information as the algorithm converges. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
