Independent component analysis: A new possibility for analysing series of electron energy loss spectra

Bonnet, N.; Nuzillard, Danielle

doi:10.1016/j.ultramic.2004.11.003

Cited by 52 publications

(28 citation statements)

References 24 publications

(37 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…3 and 4, in which the z direction corresponds to the growth axis and the x direction is parallel to the interface. Spectra have been recorded along ten lines parallel to the interface, and principal component analysis ͑PCA͒ has been applied to reduce the noise 35. Figure 3displays the two-dimensional ͑2D͒ elemental cartography.…”

mentioning

confidence: 99%

TEM and EELS measurements of interface roughness in epitaxial Fe/MgO/Fe magnetic tunnel junctions

et al. 2009

View full text Add to dashboard Cite

The crystalline structure, chemical composition, and bonding states across epitaxial Fe/MgO/Fe͑001͒ magnetic tunnel junctions grown by molecular-beam epitaxy have been investigated down to the atomic scale by spatially resolved electron energy-loss spectroscopy. Both metal-insulator interfaces exhibit significant roughness, which can be attributed to Fe and MgO terraces overlapping one another. These terraces extend over typical widths of 6-10 nm parallel to the interface and over typical heights below 1 nm, and a structural asymmetry of the roughness is revealed. These features could be responsible for the nonsymmetrical transport properties measured when reversing the applied voltage.

show abstract

mentioning

confidence: 99%

TEM and EELS measurements of interface roughness in epitaxial Fe/MgO/Fe magnetic tunnel junctions

et al. 2009

View full text Add to dashboard Cite

show abstract

“…Dobigeon and Brun [7] compared the results obtained by applying PCA, independent component analysis (ICA) [8], vertex component analysis (VCA) [9], and Bayesian linear unmixing (BLU) [10] to experimental EELS-HSI data. They eventually found that BLU provided the most plausible spatial distributions for the constituent spectral components, presumably because of its more realistic modeling of the EELS-HSI data.…”

Section: Introductionmentioning

confidence: 99%

High Spatial Resolution Hyperspectral Imaging with Machine-Learning Techniques

Shiga

Muto

2018

Nanoinformatics

View full text Add to dashboard Cite

Recent advances in scanning transmission electron microscopy (STEM) techniques have enabled us to obtain spectroscopic datasets such as those generated by electron energy-loss (EELS)/energy-dispersive X-ray (EDX) spectroscopy measurements in a PC-controlled way from a specified region of interest (ROI) even at atomic scale resolution, also known as hyperspectral imaging (HSI). Instead of conventional analytical procedures, in which the potential constituent chemical components are manually identified and the chemical state of each spectral component is successively determined, a statistical machine-learning approach, which is known to be more effective and efficient for the automatic resolution and extraction of the underlying chemical components stored in a huge three-dimensional array of an observed HSI dataset, is used. Among the statistical approaches suitable for processing HSI datasets, methods based on matrix factorization such as principal component analysis (PCA), multivariate curve resolution (MCR), and nonnegative matrix factorization (NMF) are useful to find an essential low-dimensional data subspace hidden in the HSI dataset. This chapter describes our developed NMF method, which has two additional terms in the objective function, and which is particularly effective for analyzing STEM-EELS/EDX HSI datasets: (i) a soft orthogonal penalty, which clearly resolves partially overlapped spectral components in their spatial distributions and (ii) an automatic relevance determination (ARD) prior, which optimizes the number of components involved in the observed

show abstract

“…In the case of spectrum imaging, each (x,y) pixel can be thought of as a microphone, and the signal axis (E) as the recorded sound amplitude. ICA has been applied to EDX and EELS spectrum images. More recently, the components of ICA have been used to find the composition of the bimetallic core in a core–shell nanoparticle system and the composition of gamma' precipitates in a nickel‐based superalloy .…”

Section: Introductionmentioning

confidence: 99%

“…In the case of spectrum imaging, each (x,y) pixel can be thought of as a microphone, and the signal axis (E) as the recorded sound amplitude. ICA has been applied to EDX [ 8,9 ] and EELS [ 10,11 ] spectrum images. More…”

mentioning

confidence: 99%

A New Method for Determining the Composition of Core–Shell Nanoparticles via Dual‐EDX+EELS Spectrum Imaging

Rossouw

Knappett

Wheatley

et al. 2016

Part & Part Syst Charact

View full text Add to dashboard Cite

Simultaneously acquired microanalytical X‐ray and electron energy loss signals are obtained from a bimetallic core–shell nanoparticle system (FePt@Fe3O4). The signals are decomposed using independent component analysis and the extracted components are used to separately quantify the composition of the spatially overlapping core and shell phases in the nanoheterostructure. The utilization of the complementary strengths of energy dispersive X‐ray and electron energy‐loss spectroscopy microanalysis has enabled the quantification of both light and heavy elements in a single spectrum image acquisition.

show abstract

Independent component analysis: A new possibility for analysing series of electron energy loss spectra

Cited by 52 publications

References 24 publications

TEM and EELS measurements of interface roughness in epitaxial Fe/MgO/Fe magnetic tunnel junctions

TEM and EELS measurements of interface roughness in epitaxial Fe/MgO/Fe magnetic tunnel junctions

High Spatial Resolution Hyperspectral Imaging with Machine-Learning Techniques

A New Method for Determining the Composition of Core–Shell Nanoparticles via Dual‐EDX+EELS Spectrum Imaging

Contact Info

Product

Resources

About