Search results for "feature"
showing 10 items of 4091 documents
Diagnosis of inverter-fed induction motors in short time windows using physics-assisted deep learning framework
2019
This article presents a framework for accurate fault diagnostics in inverter-fed induction machinery operating under variable speed and load conditions within very short time windows. Condition indicators based on fault characteristic frequencies observed over the extended Park's vector modulus are fused with deep features extracted using stacked autoencoders to generate a multidimensional feature space for fault classification using support vector machine. The proposed approach is demonstrated in a laboratory setup to detect the most commonly occurring faults, namely, the stator turns fault, broken rotor bars fault and bearing fault with an accuracy > 98% within a short time window of 2–3 …
Definition and Performance Evaluation of a Robust SVM Based Fall Detection Solution
2012
We propose an automatic approach to detect falls in home environment. A Support Vector Machine based classifier is fed by a set of selected features extracted from human body silhouette tracking. The classifier is followed by filtering operations taking into account the temporal nature of a video. The features are based on height and width of human body bounding box, the user's trajectory with her/his orientation, Projection Histograms and moments of order 0, 1 and 2. We study several combinations of usual transformations of the features (Fourier Transform, Wavelet transform, first and second derivatives), and we show experimentally that it is possible to achieve high performance using a si…
Experiments in Value Function Approximation with Sparse Support Vector Regression
2004
We present first experiments using Support Vector Regression as function approximator for an on-line, sarsa-like reinforcement learner. To overcome the batch nature of SVR two ideas are employed. The first is sparse greedy approximation: the data is projected onto the subspace spanned by only a small subset of the original data (in feature space). This subset can be built up in an on-line fashion. Second, we use the sparsified data to solve a reduced quadratic problem, where the number of variables is independent of the total number of training samples seen. The feasability of this approach is demonstrated on two common toy-problems.
Real-time flaw detection on complex part: Study of SVM and hyperrectangle based method
2002
We present in this paper the study of two classifications methods used in order to control in real-time some industrials parts. We present the practical frame in which is made the operations, natures of the anomaly to be detected as well as the features extractions method. We tested two techniques of classification, with different algorithm complexities and performances. We compare the results obtained on various features spaces. We end by a combinatorial perspective of results of classification.
A numerical simulation scheme for the albedo of city street canyons
1985
A numerical scheme is described for the calculation of effective albedo values of long city street canyons. The method is based on a generalization of the radiation model for inclined surfaces recently presented by Bruhl and Zdunkowski (1983). Calculated albedo values are compared with Aida's (1982) experimentally determined results. It is found that experiment and theory are in reasonable and in some cases in excellent agreement. Additional results obtained by varying the geometry of the street canyon as well as the surface reflectivities are shown to demonstrate the versatility of the calculation scheme.
Interacción de procesos geomórficos y distribución de componentes de la superficie del suelo en relación a la evolución de los abarrancamientos de Ta…
2009
Este artículo se ha desarrollado en el contexto de los proyectos: REN2003-045707GLO(PECOS); CGL2007-63258/BOS(PREVEA) y CGL2006-11619(PROBASE), financiados por la CICYT en el marco del Plan Nacional I+D+I
Ridge-enhanced optical transmission through a continuous metal film
2004
Optical transmission through a continuous (without holes) metal film with a periodic structure of metal or dielectric ridges on one or both interfaces was numerically studied. The dependencies of the transmission on the ridge width and height as well as the ridge arrangements on the opposite interfaces were investigated in weak- and strong-coupling regimes. The transmission enhancement was shown to depend on the relative position of the ridge gratings on the opposite interfaces of a film, confirming the role of resonant tunneling processes involving states of the surface polariton Bloch modes.
Land surface emissivity retrieval from airborne sensor over urban areas
2012
Abstract In this paper we compare three different methodologies to retrieve land surface emissivity (LSE) over urban areas: the NDVI thresholds method, the temperature and emissivity separation (TES) algorithm and the temperature independent spectral indices (TISI) algorithm. The methodologies were applied to the Airborne Hyperspectral Scanner (AHS) imagery acquired during the Dual-use European Security IR Experiment 2008 (DESIREX 2008) experimental campaign over the city of Madrid (Spain). The images have a spatial resolution of 4 m. The retrieved values are compared to in situ data measured during the campaign at 4 sites. Results show a good performance of the TISI and the TES algorithms …
Testing models of dental development in the earliest bony vertebrates, Andreolepis and Lophosteus
2012
Theories on the development and evolution of teeth have long been biased by the fallacy that chondrichthyans reflect the ancestral condition for jawed vertebrates. However, correctly resolving the nature of the primitive vertebrate dentition is challenged by a dearth of evidence on dental development in primitive osteichthyans. Jaw elements from the Silurian–Devonian stem-osteichthyansLophosteusandAndreolepishave been described to bear a dentition arranged in longitudinal rows and vertical files, reminiscent of a pattern of successional development. We tested this inference, using synchrotron radiation X-ray tomographic microscopy (SRXTM) to reveal the pattern of skeletal development preser…
Surface Activity of Humic Substances Within Peat Profile
2012
Humic substances are able to reduce the surface tension of their solutions and thus can act as surface-active substances in natural environment and possibly also for industrial applications. The ability to influence the surface tension of humic substance solutions depends on their origin. The objective of this study is a comparison of the ability of humic substances from peat profile to influence the surface tension of their solutions and identification of the structural characteristics of peat humic acids, determining their surfactant properties. Humic acids isolated from peat demonstrated significant impact of the surface tension of their solutions – they act as weak surfactants. The surf…