A Multi-Scaling Forward-Backward Time-Stepping Method for microwave imaging

Moriyama, Toshifumi; Oliveri, Giacomo; Salucci, Marco; Takenaka, Takashi

doi:10.1587/elex.11.20140578

Cited by 4 publications

(2 citation statements)

References 46 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Inverse scattering to solve unknown problems can be unravelled with deterministic [1][2][3] or stochastic [4][5][6] inversion solutions. The solution can be formulated either in frequency or time-domain.…”

Section: Introductionmentioning

confidence: 99%

“…Iterative multi-scaling approach (IMSA) is widely applied to increase the precision details of reconstructed images [13,14]. Segmentation concept in IMSA [1,15,16] to extract object location was adapted into AS-ROI method to define new and fine reconstruction region. Innovative aspect of this research work lies in the number of steps required to extract the object's area.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Automated Scaling Region of Interest With Iterative Edge Preserving in Forward-Backward Time-Stepping

Nawawi¹,

Sahrani²,

Ping³

2018

PIER M

View full text Add to dashboard Cite

A one-shot rescaling process, namely Automated Scaling Region of Interest (AS-ROI), is combined with an inversion technique of Forward-Backward Time-Stepping (FBTS). The purpose is to alleviate the ill-posedness and nonlinearity of inverse problem by reducing the size of the unknown problem. The inversion solution is carried out to reconstruct tumour as an unknown object in coarse investigation domain of lung area which is then rescaled down corresponding to object location and size. In this paper, edge preserving methods consisting of edge preserving regularization and anisotropic diffusion are imposed alternately on the solution and reconstructed profiles to improve the current method of AS-ROI. Results on the reconstructed lungs and tumours give significant insight of the proposed work. Accuracy level for the reconstructed profiles are significantly improved in spite that spatial resolution is retained as the original setting of FBTS.

show abstract