An automated mineral classifier using Raman spectra

Ishikawa, Sascha T.; Gulick, V. C.

doi:10.1016/j.cageo.2013.01.011

Cited by 66 publications

(49 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Although PCA preprocessing was shown to improve accuracy in a previous study, [22] our results suggest the opposite (Fig. 6).…”

Section: Effect Of Preprocessingcontrasting

confidence: 52%

“…Results of this study can be compared against those of Ishikawa and Gulick, [22] who used a much smaller subset of minerals from the RRUFF data set. Their paper lists the mineral species included in their data set but does not specify that individual spectra were used.…”

Section: Comparison With a Smaller Reference Data Setmentioning

confidence: 99%

“…Some approaches restrict the task to identification of a specific component, [14,15] while others attempt to cluster spectra into logical groups. [19,22] Most methods employ some combination of spectrum preprocessing steps to reduce the influence of noise and fluorescence, [23,24] and some also project spectra into a lower-dimensional feature space, typically using principal components analysis (PCA). [22,25] These applications focus on single-phase identification, although some workers have proposed algorithms for determining the relative contributions of phase in simple mixtures.…”

Section: Introductionmentioning

confidence: 99%

“…[19,22] Most methods employ some combination of spectrum preprocessing steps to reduce the influence of noise and fluorescence, [23,24] and some also project spectra into a lower-dimensional feature space, typically using principal components analysis (PCA). [22,25] These applications focus on single-phase identification, although some workers have proposed algorithms for determining the relative contributions of phase in simple mixtures. [26][27][28] Specifically in the geosciences, pioneering work in automated mineral identification has been performed by Pablo Sobron and colleagues involved with the RLS instrument on ExoMars, [6,18,26] building upon earlier technique development at Washington University by Larry Haskin and Alian Wang.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Machine learning tools formineral recognition and classification from Raman spectroscopy

Carey

Boucher

Mahadevan

et al. 2015

J Raman Spectroscopy

View full text Add to dashboard Cite

Tools for mineral identification based on Raman spectroscopy fall into two groups: those that are largely based on fits to diagnostic peaks associated with specific phases, and those that use the entire spectral range for multivariate analyses. In this project, we apply machine learning techniques to improve mineral identification using the latter group. We test the effects of common spectrum preprocessing steps, such as intensity normalization, smoothing, and squashing, and found that the last is superior. Next, we demonstrate that full-spectrum matching algorithms exhibit excellent performance in classification tasks, without requiring time-intensive dimensionality reduction or model training. This class of algorithms supports both vector and trajectory input formats, exploiting all available spectral information. By combining these insights, we find that optimal mineral spectrum matching performance can be achieved using careful preprocessing and a weighted-neighbors classifier based on a vector similarity metric. IntroductionUse of Raman spectroscopy in the geosciences is growing rapidly, as evidenced by burgeoning publications in the fields of geology, cultural anthropology, environmental science, and, in particular, planetary science. Raman spectrometers have been proposed for exploration of a diverse range of extraterrestrial targets including asteroids, [1] Europa, [2,3] Mars, [4][5][6] the Moon, [7] and Venus. [8] A Raman laser spectrometer (RLS) is part of the science instrument payload of the European Space Agency 2018 ExoMars mission; the RLS instrument will target mineralogical and astrobiological investigations on the surface and subsurface of Mars. [9,10] Raman will also be used on the upcoming NASA Mars 2020 mission as part of the SuperCam and SHERLOC instruments. [11] What all these applications have in common is their dependence on software and mineralogical databases for phase identification and quantification of relative abundances of mineral components. Because of the structural diversity and chemical complexity of naturally occurring minerals, optimal applications for these purposes require an infusion of work into development of appropriate software and mineral databases.In practice, users commonly depend on a combination of matching software distributed by spectrometer manufacturers and one-on-one comparisons with database spectra for their identifications, but these types of identification have two critical limitations. First, identifications are only as good as the databases used to match them. No database can be entirely comprehensive, so it cannot be unequivocally claimed that any spectrum is a 'perfect match' to exactly one other phase.The RRUFF database was founded in 2006 at Arizona State University by Robert Downs to remedy this situation by providing coverage of all known mineral species. [12] It has quickly become the preeminent resource available for Raman spectra of minerals. RRUFF currently contains over 20 378 spectra acquired at several different laser wavelengths from orien...

show abstract

“…Although PCA preprocessing was shown to improve accuracy in a previous study, [22] our results suggest the opposite (Fig. 6).…”

Section: Effect Of Preprocessingcontrasting

confidence: 52%

Section: Comparison With a Smaller Reference Data Setmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Machine learning tools formineral recognition and classification from Raman spectroscopy

Carey

Boucher

Mahadevan

et al. 2015

J Raman Spectroscopy

View full text Add to dashboard Cite

show abstract

“…SVM has high generalization performance using a hyperplane, providing localized and global solutions, while conventional neural networks often converge on only the local minima. Its feasibility for chemical analysis was previously reported, e.g., for identifying components or structure such as mass spectra (Bern et al, 2004;Hilario et al, 2006;Somorjai et al, 2003), FT-IR spectra (Ferrão et al, 2007), NIR spectra (Chauchard et al, 2004;Chen et al, 2006;Devos et al, 2009), and Raman spectra (Gaus et al, 2006;Ishikawa and Gulick, 2013;Rösch et al, 2005;Sattlecker et al, 2010). However, its feasibility for analyzing radiation spectra is yet to be studied.…”

Section: Introductionmentioning

confidence: 96%

Application of support vector machine to rapid classification of uranium waste drums using low-resolution γ-ray spectra

Hata

Yokoyama

Ishimori

et al. 2015

Applied Radiation and Isotopes

View full text Add to dashboard Cite

Mineralogical investigations using XRD, XRF, and Raman spectroscopy in a combined approach

Secchi

Zanatta

Borovin

et al. 2018

J Raman Spectroscopy

View full text Add to dashboard Cite

The identification of mineralogical phases in drill cores is one of the most challenging tasks in the mining activity in view of an efficient metal extraction. This process requires the analytical characterization of large volumes of material to obtain a complete set of data in a minimum of time. None of the commonly used methods in mineralogical analyses, such as IR‐based techniques, X‐rays fluorescence, and hyperspectral imaging, is capable to provide a fully satisfactory response for several reasons, the main one being the complexity of the ores. Moreover, the characterization is often conducted in remote laboratories and only on selected samples to limit the time waste. A possible alternative solution requires a multianalytical approach exploiting on‐field techniques. This strategy is currently being developed within SOLSA, a joint EU H2020 project, and consists of an automatic expert system coupling sonic drilling, imaging, profilometer, hyperspectral cameras, and a combination of Raman spectroscopy, X‐rays fluorescence, and X‐rays diffraction. In this work, the principles on which this cooperative approach is based are discussed, with application to two specific test samples, showing the potential and novelty of the method. In particular, a case is considered in which the sample characterization by the separate use of a single technique fails, whereas the combination of the three analyses (Raman spectroscopy, X‐rays fluorescence, and X‐rays diffraction) works even if the system is very complex.

show abstract

An automated mineral classifier using Raman spectra

Cited by 66 publications

References 17 publications

Machine learning tools formineral recognition and classification from Raman spectroscopy

Machine learning tools formineral recognition and classification from Raman spectroscopy

Application of support vector machine to rapid classification of uranium waste drums using low-resolution γ-ray spectra

Mineralogical investigations using XRD, XRF, and Raman spectroscopy in a combined approach

Contact Info

Product

Resources

About