A Comparison of Some Methods for the Selection of a Common Eigenvector Model for the Covariance Matrices of Two Groups

2014 ◽  
Vol 45 (8) ◽  
pp. 2917-2936 ◽  
Author(s):  
P. T. Pepler ◽  
D. W. Uys ◽  
D. G. Nel
2008 ◽  
Vol 15 (2) ◽  
pp. 27-42
Author(s):  
Denis Altieri De Oliveira Moraes ◽  
Victor Haertel

This paper deals with the problem of classifying high-dimensional image data image data using a multiple stage classifier structured as a binary tree. The aim here consists in finding the optimal structure for the binary tree in the sense of achieving a stable accuracy. The advantage presented by a multiple stage classifier lies on the fact that only a sub-set of classes is considered at each stage, allowing a better selection of the features to be used at each node.  The binary tree is a particular case of a tree structured classifier, on which only two classes are considered at each node. This peculiarity makes possible the direct use of statistical distances for feature reduction (selection or extraction). In this study the criterion used for feature reduction at each node consists in optimizing the Bhattacharyya distance separating both classes in the node. The optimization of Bhattacharyya distance was based on the covariance matrices. Once the final set of features is obtained at each particular node, the classification is performed using the Gaussian Maximum Likelihood decision rule. Tests were performed using high-dimensional image data collected by the sensor system AVIRIS covering a test area. The criteria to evaluate the performance of the classifiers are: the final accuracy yielded by the classifier, its stability, and the required computer time.


2017 ◽  
Vol 29 (6) ◽  
pp. 1631-1666 ◽  
Author(s):  
Takashi Uehara ◽  
Matteo Sartori ◽  
Toshihisa Tanaka ◽  
Simone Fiori

The estimation of covariance matrices is of prime importance to analyze the distribution of multivariate signals. In motor imagery–based brain-computer interfaces (MI-BCI), covariance matrices play a central role in the extraction of features from recorded electroencephalograms (EEGs); therefore, correctly estimating covariance is crucial for EEG classification. This letter discusses algorithms to average sample covariance matrices (SCMs) for the selection of the reference matrix in tangent space mapping (TSM)–based MI-BCI. Tangent space mapping is a powerful method of feature extraction and strongly depends on the selection of a reference covariance matrix. In general, the observed signals may include outliers; therefore, taking the geometric mean of SCMs as the reference matrix may not be the best choice. In order to deal with the effects of outliers, robust estimators have to be used. In particular, we discuss and test the use of geometric medians and trimmed averages (defined on the basis of several metrics) as robust estimators. The main idea behind trimmed averages is to eliminate data that exhibit the largest distance from the average covariance calculated on the basis of all available data. The results of the experiments show that while the geometric medians show little differences from conventional methods in terms of classification accuracy in the classification of electroencephalographic recordings, the trimmed averages show significant improvement for all subjects.


2019 ◽  
Vol 42 ◽  
Author(s):  
Gian Domenico Iannetti ◽  
Giorgio Vallortigara

Abstract Some of the foundations of Heyes’ radical reasoning seem to be based on a fractional selection of available evidence. Using an ethological perspective, we argue against Heyes’ rapid dismissal of innate cognitive instincts. Heyes’ use of fMRI studies of literacy to claim that culture assembles pieces of mental technology seems an example of incorrect reverse inferences and overlap theories pervasive in cognitive neuroscience.


1975 ◽  
Vol 26 ◽  
pp. 395-407
Author(s):  
S. Henriksen

The first question to be answered, in seeking coordinate systems for geodynamics, is: what is geodynamics? The answer is, of course, that geodynamics is that part of geophysics which is concerned with movements of the Earth, as opposed to geostatics which is the physics of the stationary Earth. But as far as we know, there is no stationary Earth – epur sic monere. So geodynamics is actually coextensive with geophysics, and coordinate systems suitable for the one should be suitable for the other. At the present time, there are not many coordinate systems, if any, that can be identified with a static Earth. Certainly the only coordinate of aeronomic (atmospheric) interest is the height, and this is usually either as geodynamic height or as pressure. In oceanology, the most important coordinate is depth, and this, like heights in the atmosphere, is expressed as metric depth from mean sea level, as geodynamic depth, or as pressure. Only for the earth do we find “static” systems in use, ana even here there is real question as to whether the systems are dynamic or static. So it would seem that our answer to the question, of what kind, of coordinate systems are we seeking, must be that we are looking for the same systems as are used in geophysics, and these systems are dynamic in nature already – that is, their definition involvestime.


1978 ◽  
Vol 48 ◽  
pp. 515-521
Author(s):  
W. Nicholson

SummaryA routine has been developed for the processing of the 5820 plates of the survey. The plates are measured on the automatic measuring machine, GALAXY, and the measures are subsequently processed by computer, to edit and then refer them to the SAO catalogue. A start has been made on measuring the plates, but the final selection of stars to be made is still a matter for discussion.


Author(s):  
P.J. Killingworth ◽  
M. Warren

Ultimate resolution in the scanning electron microscope is determined not only by the diameter of the incident electron beam, but by interaction of that beam with the specimen material. Generally, while minimum beam diameter diminishes with increasing voltage, due to the reduced effect of aberration component and magnetic interference, the excited volume within the sample increases with electron energy. Thus, for any given material and imaging signal, there is an optimum volt age to achieve best resolution.In the case of organic materials, which are in general of low density and electric ally non-conducting; and may in addition be susceptible to radiation and heat damage, the selection of correct operating parameters is extremely critical and is achiev ed by interative adjustment.


Author(s):  
P. M. Lowrie ◽  
W. S. Tyler

The importance of examining stained 1 to 2μ plastic sections by light microscopy has long been recognized, both for increased definition of many histologic features and for selection of specimen samples to be used in ultrastructural studies. Selection of specimens with specific orien ation relative to anatomical structures becomes of critical importance in ultrastructural investigations of organs such as the lung. The uantity of blocks necessary to locate special areas of interest by random sampling is large, however, and the method is lacking in precision. Several methods have been described for selection of specific areas for electron microscopy using light microscopic evaluation of paraffin, epoxy-infiltrated, or epoxy-embedded large blocks from which thick sections were cut. Selected areas from these thick sections were subsequently removed and re-embedded or attached to blank precasted blocks and resectioned for transmission electron microscopy (TEM).


Author(s):  
K.-H. Herrmann ◽  
D. Krahl ◽  
H.-P Rust

The high detection quantum efficiency (DQE) is the main requirement for an imagerecording system used in electron microscopy of radiation-sensitive specimens. An electronic TV system of the type shown in Fig. 1 fulfills these conditions and can be used for either analog or digital image storage and processing [1], Several sources of noise may reduce the DQE, and therefore a careful selection of various elements is imperative.The noise of target and of video amplifier can be neglected when the converter stages produce sufficient target electrons per incident primary electron. The required gain depends on the type of the tube and also on the type of the signal processing chosen. For EBS tubes, for example, it exceeds 10. The ideal case, in which all impinging electrons create uniform charge peaks at the target, is not obtainable for several reasons, and these will be discussed as they relate to a system with a scintillator, fiber-optic and photo-cathode combination as the first stage.


Sign in / Sign up

Export Citation Format

Share Document