Learn More
A new algorithm for extracting features from images for object recognition is described. The algorithm uses higher order spectra to provide desirable invariance properties, to provide noise immunity, and to incorporate nonlinearity into the feature extraction procedure thereby allowing the use of simple classifiers. An image can be reduced to a set of 1D(More)
An exploratory study of the bispectrum of economic time series, " Appl. Bispectra of ocean waves in time series analysis, " ed. by M. Rosenblatt, pp. 125-139, 1963. M. J. Hinich, " Testing for Gaussianity and linearity of a stationary time series, " J. Real time bispectral analysis of Gear noise and its applications to contactless diagnosis, ". I. A test(More)
The detection of seizure in the newborn is a critical aspect of neurological research. Current automatic detection techniques are difficult to assess due to the problems associated with acquiring and labelling newborn electroencephalogram (EEG) data. A realistic model for newborn EEG would allow confident development, assessment and comparison of these(More)
—This paper presents the essential elements for developing objective methods of assessment of the performance of time-frequency signal analysis techniques. We define a measure for assessing the resolution performance of time-frequency distributions (TFDs) in separating closely spaced components in the time-frequency domain. The measure takes into account(More)
— This paper deals with the problem of blind separation of nonstationary sources in the underdetermined case, i.e. more sources than sensors, using time–frequency distributions (TFDs). We propose a new algorithm to achieve the separation based on a main assumption of time–frequency (TF) disjoint sources that allows an explicit exploitation of(More)
The Wigner-Ville distribution (WVD) is a valuable tool for time-frequency signal analysis. In order to implement the WVD in real time, an efficient algorithm and architecture have been developed which may be implemented with-commercial components. This algorithm successively computes the analytic signal corresponding to the input signal, forms a weighted(More)