Search results for " density"

Semisupervised kernel orthonormalized partial least squares

2012

This paper presents a semisupervised kernel orthonormalized partial least squares (SS-KOPLS) algorithm for non-linear feature extraction. The proposed method finds projections that minimize the least squares regression error in Hilbert spaces and incorporates the wealth of unlabeled information to deal with small size labeled datasets. The method relies on combining a standard RBF kernel using labeled information, and a generative kernel learned by clustering all available data. The positive definiteness of the kernels is proven, and the structure and information content of the derived kernels is studied. The effectiveness of the proposed method is successfully illustrated in standard UCI d…

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 measurement-based trajectory model for drifted motions towards a target zone

2016

Trajectory models have numerous applications in the area of wiewlwss communications. The aim of this paper is to develop an empirical trajectory model for drifted motions. Recently, a highly flexible trajectory model based on the primitives of Brownian fields (TramBrown) was proposed by A. Borhani and M. Patzold. This paper provides an empirical proof for TramBrown using global positioning system (GPS) data collected from real life user traces drifting to a particular target point or a zone. The recorded location coordinates of the mobile user are processed to compute the total travelling length and the angle-of-motion (AOM) along the drifted trajectory. It is shown that the probability den…

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.

Effects of feeding frequency on growth and food utilisation of rainbow trout (Oncorhynchus mykiss) fed low-fat herring or dry pellets

1998

Abstract One-year-old rainbow trout (weight range 400–700 g) were fed for 18 weeks on low-fat herring or commercial dry pellets 1, 2 or 4 times in a day. Quadratic regression analysis indicated that at least three feedings were required for maximum growth and that fish fed a dry diet could benefit from even more frequent feeding. The proportion of lipid in growth increased with increased number of feedings, but the protein content was not affected (mean 18.6% protein wet basis). Food consumption (dry weight) was affected in a similar manner for both diets and its maximum occurred at slightly higher frequency than for maximum growth. No effects of feeding frequency or type of diet on energy …

Optimization of the straightness measurements on rough surfaces by Monte Carlo simulation

2013

Summary The straightness error of a coordinate measuring machine (CMM) is determined by measuring a rule standard. Thanks to a reversal technique, the straightness uncertainty of the CMM is theoretically dissociated from the straightness uncertainty of the rule. However, stochastic variations of the whole measurement system involve uncertainties of the CMM straightness error. To quantify these uncertainties, different sources of stochastic variations are listed with their associated probability density functions. Then Monte Carlo methods are performed first to quantify error and secondly to optimize measurement protocol. It is shown that a 5-measurement distance from 0.1 mm to each measurem…

Frequency composition of traction and tillage forces on a mole plough

1997

Field experiments were carried out in order to quantify the variations of the traction and tillage forces on a mole plough, in the case of four clayey soils of Eastern France. The same tractor-implement system, the same forward working speed (1 m s−1) and the same length of ploughing (20 m) were used. Specially designed strain gauge transducers applied to the system delivered a signal which was numerised and analysed in terms of power spectral density (PSD) function. The frequency composition of the traction and tillage forces consisted of interactions between soil and tool as well as vibrations of the tractor-implement system. Low frequencies (0–2 Hz range) were related to the random prope…

The impact of the age of vines on soil hydraulic conductivity in vineyards in eastern Spain

2017

Soil infiltration processes manage runoff generation, which in turn affects soil erosion. There is limited information on infiltration rates. In this study, the impact of vine age on soil bulk density (BD) and hydraulic conductivity (Ks) was assessed on a loam soil tilled by chisel plough. Soil sampling was conducted in the inter row area of six vineyards, which differed by the age from planting: 0 (Age 0; just planted), 1, 3, 6, 13, and 25 years (Age 1, Age 3, Age 6, Age 13, and Age 25, respectively). The One Ponding Depth (OPD) approach was applied to ring infiltration data to estimate soil Ks with an α* parameter equal to 0.012 mm-1. Soil bulk density for Age 0 was about 1.5 times greate…

Characterization of defect density states in MoOx for c-Si solar cell applications

Layers of MoOx have been deposited by thermal evaporation followed by post-deposition annealing (PDA). The density of states (DOS) distributions of the MoOx films were extracted deconvoluting the absorption spectra, measured by a phothermal deflection spectroscopy setup, including the small polaron contribution. Results revealed a sub-band defect distribution centered 1.1 eV below the conduction band; the amplitude of this distribution was found to increase with PDA temperature and film thickness.