Data fusion and dual-domain classification analysis of pigments studied in works of art

Ramos, Pablo Manuel; Ruisánchez, Itziar

doi:10.1016/j.aca.2005.10.066

Cited by 27 publications

(14 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Fusion of the Data: Proposed Methodology. Ramos et al 2,17 suggest two different strategies to fuse spectral data: data level fusion or low-level fusion on the one hand and feature level fusion or mid-level fusion on the other hand. When using low-level fusion the raw spectral data are fused, whereas when using mid-level fusion the fusion is done ideally over the most relevant features, so variable selection is performed before the fusion.…”

Section: Resultsmentioning

confidence: 99%

“…Raman spectroscopy and X-ray fluorescence (XRF) spectroscopy are particularly well suited for the analysis of ancient art objects in cultural heritage studies because both techniques are fast, sensitive, and noninvasive and measurements can take place in situ. 1,2 Raman spectroscopy provides molecular information about the material, while XRF spectroscopy provides the elementary composition. Moreover, the penetration depth of the X-rays is larger than the penetration depth of the laser light, resulting in combined information of different layers.…”

Section: Introductionmentioning

confidence: 99%

“…The aim of the current research is to merge the spectra and process the fused data together. Ramos et al 2,17 proposed a mid-level data fusion system to process Raman and XRF spectra obtained from a combined micro-Raman-XRF instrument. In the first step of their data fusion system wavelet transformation is used to split each spectrum set into blocks, according to their frequency.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Identification of Inorganic Pigments Used in Porcelain Cards Based on Fusing Raman and X-ray Fluorescence (XRF) Data

et al. 2011

View full text Add to dashboard Cite

Raman spectroscopy and X-ray fluorescence (XRF) spectroscopy are often used as complementary techniques that are well suited for the analysis of art objects because both techniques are fast, sensitive, and noninvasive and measurements can take place in situ. In most of these studies, both techniques are used separately, in the sense that the spectra are evaluated independently and single conclusions are obtained, considering both results. This paper presents a data fusion procedure for Raman and XRF data for the characterization of pigments used in porcelain cards. For the classification of the analyzed points of the porcelain cards principal component analysis (PCA) was used. A first attempt was made to develop a new procedure for the identification of the pigments using a database containing the fused Raman-XRF data of 24 reference pigments. The results show that the classification based on the fused Raman-XRF data is significantly better than the classifications based on the Raman data or the XRF data separately.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Identification of Inorganic Pigments Used in Porcelain Cards Based on Fusing Raman and X-ray Fluorescence (XRF) Data

et al. 2011

View full text Add to dashboard Cite

show abstract

“…Formal statistical explanation has been given to show the advantage of using PLS for discriminant analysis. [25][26][27][28][29][30][31][32] After mean centering, it was found that 85.0% of the spectral (x-block) variance was described by the first four PLS factors, while 98.1% of the target matrix (y-block) variance was described.…”

Section: Methodsmentioning

confidence: 99%

Drug Characterization in Low Dosage Pharmaceutical Tablets Using Raman Microscopic Mapping

Henson

Zhang

2006

Appl Spectrosc

View full text Add to dashboard Cite

Raman micro-spectroscopic mapping is utilized to analyze pharmaceutical tablets containing a low concentration (0.5% w/w) of active pharmaceutical ingredient (API). The domain sizes and spatial distributions of the API and the major excipients are obtained. Domain size of the API is found to be dependent upon the particle size distribution of the ingoing API material, making the Raman maps good indicators of the source of API used in tablet manufacturing. Multivariate classification was performed to simultaneously check for the presence of two undesired API polymorphs within tablets. Raman mapping was demonstrated capable of detecting in the tablet matrix as little as 10% form conversion of the low-dosage (0.5% w/w) API, which is equivalent to detection of a 0.05% w/w polymorphic impurity. Overall, the information provided by Raman micro-spectroscopic mapping was found to have potential utility for manufacturing process optimization or predictive stability assessments.

show abstract

“…Broadly speaking, most of the previous work in data fusion may be categorized into low-level (sensorlevel) fusion [12] [38], mid-level fusion [31], and high-level (decision-level) fusion [35], [5].…”

Section: Multi-view Data Fusionmentioning

confidence: 99%

A visualization framework for team sports captured using multiple static cameras

Hamid

Kumar

Hodgins

et al. 2014

Computer Vision and Image Understanding

View full text Add to dashboard Cite

Abstract-We present a novel approach for robust localization of multiple people observed using a set of static cameras. We use this location information to generate a visualization of the virtual offside line in soccer games. To compute the position of the offside line, we need to localize players positions, and identify their team roles. We solve the problem of fusing corresponding players positional information by finding minimum weight K-length cycles in a complete K-partite graph. Each partite of the graph corresponds to one of the K cameras, whereas each node of a partite encodes the position and appearance of a player observed from a particular camera. To find the minimum weight cycles in this graph, we use a dynamic programming based approach that varies over a continuum from maximally to minimally greedy in terms of the number of graph-paths explored at each iteration. We present proofs for the efficiency and performance bounds of our algorithms. Finally, we demonstrate the robustness of our framework by testing it on 82,000 frames of soccer footage captured over eight different illumination conditions, play types, and team attire. Our framework runs in near-real time, and processes video from 3 full HD cameras in about 0.4 seconds for each set of corresponding 3 frames.

show abstract

Data fusion and dual-domain classification analysis of pigments studied in works of art

Cited by 27 publications

References 30 publications

Identification of Inorganic Pigments Used in Porcelain Cards Based on Fusing Raman and X-ray Fluorescence (XRF) Data

Identification of Inorganic Pigments Used in Porcelain Cards Based on Fusing Raman and X-ray Fluorescence (XRF) Data

Drug Characterization in Low Dosage Pharmaceutical Tablets Using Raman Microscopic Mapping

A visualization framework for team sports captured using multiple static cameras

Contact Info

Product

Resources

About