Search results for "Principal component"
showing 10 items of 514 documents
Comprehensive Strategy for Proton Chemical Shift Prediction: Linear Prediction with Nonlinear Corrections
2014
A fast 3D/4D structure-sensitive procedure was developed and assessed for the chemical shift prediction of protons bonded to sp3carbons, which poses the maybe greatest challenge in the NMR spectral parameter prediction. The LPNC (Linear Prediction with Nonlinear Corrections) approach combines three well-established multivariate methods viz. the principal component regression (PCR), the random forest (RF) algorithm, and the k nearest neighbors (kNN) method. The role of RF is to find nonlinear corrections for the PCR predicted shifts, while kNN is used to take full advantage of similar chemical environments. Two basic molecular models were also compared and discussed: in the MC model the desc…
Robustness of texture parameters for color texture analysis
2006
This article proposes to deal with noisy and variable size color textures. It also proposes to deal with quantization methods and to see how such methods change final results. The method we use to analyze the robustness of the textures consists of an auto-classification of modified textures. Texture parameters are computed for a set of original texture samples and stored into a database. Such a database is created for each quantization method. Textures from the set of original samples are then modified, eventually quantized and classified according to classes determined from a precomputed database. A classification is considered incorrect if the original texture is not retrieved. This metho…
Semisupervised Kernel Feature Extraction for Remote Sensing Image Analysis
2014
This paper presents a novel semisupervised kernel partial least squares (KPLS) algorithm for nonlinear feature extraction to tackle both land-cover classification and biophysical parameter retrieval problems. The proposed method finds projections of the original input data that align with the target variable (labels) and incorporates the wealth of unlabeled information to deal with low-sized or underrepresented data sets. The method relies on combining two kernel functions: the standard radial-basis-function kernel based on labeled information and a generative, i.e., probabilistic, kernel directly learned by clustering the data many times and at different scales across the data manifold. Th…
A family of kernel anomaly change detectors
2014
This paper introduces the nonlinear extension of the anomaly change detection algorithms in [1] based on the theory of reproducing kernels. The presented methods generalize their linear counterparts, under both the Gaussian and elliptically-contoured assumptions, and produce both improved detection accuracies and reduced false alarm rates. We study the Gaussianity of the data in Hilbert spaces with kernel dependence estimates, provide low-rank kernel versions to cope with the high computational cost of the methods, and give prescriptions about the selection of the kernel functions and their parameters. We illustrate the performance of the introduced kernel methods in both pervasive and anom…
Semi-Supervised Remote Sensing Image Classification based on Clustering and the Mean Map Kernel
2008
This paper presents a semi-supervised classifier based on the combination of the expectation-maximization (EM) algorithm for Gaussian mixture models (GMM) and the mean map kernel. The proposed method uses the most reliable samples in terms of maximum likelihood to compute a kernel function that accurately reflects the similarity between clusters in the kernel space. The proposed method improves classification accuracy in situations where the available labeled information does not properly describe the classes in the test image.
An Efficient Method for the Visualization of Spectral Images Based on a Perception-Oriented Spectrum Segmentation
2010
We propose a new method for the visualization of spectral images. It involves a perception-based spectrum segmentation using an adaptable thresholding of the stretched CIE standard observer colormatching functions. This allows for an underlying removal of irrelevant channels, and, consequently, an alleviation of the computational burden of further processings. Principal Components Analysis is then used in each of the three segments to extract the Red, Green and Blue primaries for final visualization. A comparison framework using two different datasets shows the efficiency of the proposed method.
Natural oxygenation of Champagne wine during ageing on lees: A metabolomics picture of hormesis
2016
International audience; The oxygenation of Champagne wine after 4 and 6 years of aging on lees in bottle was investigated by FTICR-MS and UPLC-Q-TOF-MS. Three levels of permeability were considered for the stoppers, ranging from 0.2 to 1.8 mg/L/year of oxygen transfer rate. Our results confirmed a good repeatability of ultrahigh resolution FTICR-MS, both in terms of m/z and coefficient of variation of peak intensities among biological replicates. Vintages appeared to be the most discriminated features, and metabolite annotations suggested that the oldest wines (2006) were characterized by a higher sensitivity towards oxygenation. Within each vintage, the oxygenation mechanisms appeared to b…
Polychlorinated Biphenyls in Sediments from Sicilian Coastal Area (Scoglitti) using Automated Soxhlet, GC-MS, and Principal Component Analysis
2014
A methodology for the PAHs and PCBs congener determination in sediment samples has been revised. We determined the distributions of PAHs and PCBs in the superficial sediments of the Scoglitti (Italy) coastal area to provide data for comparison with other marine systems and to hypothesize the sources. Extraction yield, for PCB, was never less than 60% in most cases, while for PAHs, utilizing perdeuterated surrogate standard (benz[a]anthracene-d12 and anthracene-d10) was never less than 72%. The total concentration of the 16 PAHs investigated, expressed as the sum of concentrations, PAHs, varied from 1–5087 μg/kg of dry matrix, while the PCBs ranged from detection limit to 36 μg/kg of dry mat…
Applying fully tensorial ICA to fMRI data
2016
There are two aspects in functional magnetic resonance imaging (fMRI) data that make them awkward to analyse with traditional multivariate methods - high order and high dimension. The first of these refers to the tensorial nature of observations as array-valued elements instead of vectors. Although this can be circumvented by vectorizing the array, doing so simultaneously loses all the structural information in the original observations. The second aspect refers to the high dimensionality along each dimension making the concept of dimension reduction a valuable tool in the processing of fMRI data. Different methods of tensor dimension reduction are currently gaining popUlarity in literature…
Identifying critical factors for implementing good agricultural practice
2009
En este artículo se presenta la identificación de los factores críticos (FC) que afectan la implantación de un programa de buenas prácticas agrícolas (BPA) en productores de café y frutas del departamento del Huila, en Colombia, mediante la realización de un análisis factorial exploratorio utilizando como método de factorización el análisis de componentes principales (ACP); las matrices de datos se construyeron con los resultados de la aplicación de sendos instrumentos con estructura definida en las dos poblaciones objeto de estudio, el instrumento Starbucks C.A.F.E. Practices -para pequeños caficultores en el caso de los productores de café- y EUREPGAP V2.1 Oct.2004 - Checklist-listado de …