Superresolved image reconstruction from incomplete data

Chuang, Yi-Chen; Dudley, Richard; Fiddy, Michael A.

doi:10.1117/12.930836

Cited by 4 publications

(3 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Propagating and evanescent waves are well models by the first Born approximation, kVa <<1, for sufficiently thin or weakly scattering objects, but not otherwise, [12]. Here V is the maximum value for the index difference, a is a measure of the dimensions of the object and k = k j > k 0 where k 0 is the free space wavenumber, 2π/λ 0 .…”

Section: Compressive Sampling Theorymentioning

confidence: 99%

See 1 more Smart Citation

Compressive sampling methods for superresolution imaging

Chuang

Dudley

Fiddy

2012

Image Reconstruction From Incomplete Data VII

Self Cite

View full text Add to dashboard Cite

We investigate superresolution imaging using negative index metamaterials. Measurement of subwavelength scale features in the image domain is tedious and compressive sampling techniques are considered to alleviate this problem. A single detector (c.f. a single pixel camera geometry) is considered from which a high resolution image can be computed, which makes use of structured illumination for coding.

show abstract

Section: Compressive Sampling Theorymentioning

confidence: 99%

“…This confuses the interpretation of the image and an inverse scattering algorithm needs to be applied in order to recover from the superresolved measured field, an estimate of the fine structure of the physical object. This problem is the focus of a companion paper in this meeting ( [12]). An example of this problem is shown in figure 2.…”

Section: Introductionmentioning

confidence: 98%

Compressive sampling methods for superresolution imaging

Chuang

Dudley

Fiddy

2012

Image Reconstruction From Incomplete Data VII

Self Cite

View full text Add to dashboard Cite

show abstract

“…Unfortunately, due to time constraints in data acquisition, the number of possible sampling points of the k vector is particularly small, and this leads to Gibbs artefacts (Gao andReeves 2000, Oppenheim andSchaffer 1989). This problem is reflected in (i) a serious failure of the spectral spatial distribution displacement, and (ii) a drastic reduction of the image spatial resolution (Liang 1989, Liang andHaacke 1990).…”

Section: Introductionmentioning

confidence: 99%

Radiofrequency field inhomogeneity compensation in high spatial resolution magnetic resonance spectroscopic imaging

2014

View full text Add to dashboard Cite

Clinical magnetic resonance spectroscopy imaging (MRSI) is a non-invasive functional technique, whose mathematical framework falls into the category of linear inverse problems. However, its use in medical diagnostics is hampered by two main problems, both linked to the Fourier-based technique usually implemented for spectra reconstruction: poor spatial resolution and severe blurring in the spatial localization of the reconstructed spectra. Moreover, the intrinsic ill-posedness of the MRSI problem might be worsened by (i) spatially dependent distortions of the static magnetic field (B0) distribution, as well as by (ii) inhomogeneity in the power deposition distribution of the radiofrequency magnetic field (B1). Among several alternative methods, slim (Spectral Localization by IMaging) and bslim (B0 compensated slim) are reconstruction algorithms in which a priori information concerning the spectroscopic target is introduced into the reconstruction kernel. Nonetheless, the influence of the B1 field, particularly when its operating wavelength is close to the size of the human organs being studied, continues to be disregarded. starslim (STAtic and Radiofrequency-compensated slim), an evolution of the slim and bslim methods, is therefore proposed, in which the transformation kernel also includes the B1 field inhomogeneity map, thus allowing almost complete 3D modelling of the MRSI problem. Moreover, an original method for the experimental determination of the B1 field inhomogeneity map specific to the target under evaluation is also included. The compensation capabilities of the proposed method have been tested and illustrated using synthetic raw data reproducing the human brain.

show abstract