Driver Maneuver Detection and Analysis Using Time Series Segmentation and Classification.

The current paper implements a methodology for automatically detecting vehicle maneuvers from vehicle telemetry data under naturalistic driving settings. Previous approaches have treated vehicle maneuver detection as a classification problem, although both time series segmentation and classification are required since input telemetry data are continuous. Our objective is to develop an end-to-end pipeline for the frame-by-frame annotation of naturalistic driving studies videos into various driving events including stop and lane-keeping events, lane changes, left-right turning movements, and horizontal curve maneuvers.

Identification of late blight resistance quantitative trait loci in Solanum pimpinellifolium accession PI 270441.

Late blight (LB), caused by the oomycete Phytophthora infestans, is one of the most destructive diseases of the cultivated tomato (Solanum lycopersicum L.) and potato (Solanum tuberosum L.) worldwide.

Predicting greenhouse gas benefits of improved nitrogen management in North American maize.

Farmers, food supply companies, and policymakers need practical yet scientifically robust methods to quantify how improved nitrogen (N) fertilizer management can reduce nitrous oxide (N O) emissions. To meet this need, we developed an empirical model based on published field data for predicting N O emission from rainfed maize (Zea mays L.) fields managed with inorganic N fertilizer in the United States and Canada.

Latin American Physical Educators' Intention to Teach Individuals With Disabilities.

Purpose: To analyze Latin American physical education (PE) teachers' intentions toward teaching students with disabilities.

Participants: 474 in-service PE teachers from 5 different Latin American countries.

Method: Descriptive survey.

Watershed land use controls on chemical inputs to lake ontario embayments.

There is considerable interest in understanding the role of land use in controlling surface water quality. This study was conducted to assess the role of land cover in regulating temporal and spatial patterns in nutrients and major solutes in rivers that drain into Lake Ontario. Water samples were collected monthly from 22 river sites in subwatersheds of eight embayments along the New York coast of Lake Ontario over the period 2001-2003.

The growth of loudness functions measured in cochlear implant listeners using absolute magnitude estimation and compared using Akaike's information criterion.

The input/output function for acoustic hearing can be characterized by the growth of loudness with sound pressure level and generally follows a compressive power law. In contrast, in electric hearing, loudness reportedly is an expansive function of applied electrical current but the specific shape of the function has not been fully determined. Loudness growth models have implications for the implementation of cochlear implant speech processors.

An evidential extension of the MRII training algorithm for detecting erroneous MADALINE responses.

This paper integrates the evidential reasoning methodology with the parallel distributed learning paradigm of artificial neural networks (ANN). As such, this work presents an algorithm for the detection and, if possible, subsequent correction of the errors in the neuron responses in the output layer of the multiple adaptive linear element (MADALINE) ANN. A geometrical perspective of the MADALINE ANN processing methodology is provided.

Efficient classification for multiclass problems using modular neural networks.

The rate of convergence of net output error is very low when training feedforward neural networks for multiclass problems using the backpropagation algorithm. While backpropagation will reduce the Euclidean distance between the actual and desired output vectors, the differences between some of the components of these vectors increase in the first iteration. Furthermore, the magnitudes of subsequent weight changes in each iteration are very small, so that many iterations are required to compensate for the increased error in some components in the initial iterations.

An improved algorithm for neural network classification of imbalanced training sets.

The backpropagation algorithm converges very slowly for two-class problems in which most of the exemplars belong to one dominant class. An analysis shows that this occurs because the computed net error gradient vector is dominated by the bigger class so much that the net error for the exemplars in the smaller class increases significantly in the initial iteration. The subsequent rate of convergence of the net error is very low.

Bounds on the number of samples needed for neural learning.

The relationship between the number of hidden nodes in a neural network, the complexity of a multiclass discrimination problem, and the number of samples needed for effect learning are discussed. Bounds for the number of samples needed for effect learning are given. It is shown that Omega(min (d,n) M) boundary samples are required for successful classification of M clusters of samples using a two-hidden-layer neural network with d-dimensional inputs and n nodes in the first hidden layer.

A quantitative comparison of attenuation-weighted backprojection with multiplicative and iterative postprocessing attenuation compensation in SPECT.

Comparisons are made between different analytic attenuation compensation methods used in SPECT imaging. The methods include a multiplicative technique and a single-iterative technique, both applied after filtered backprojection, and two different implementations of an attenuation-weighted filtered backprojection technique (A-W FBP). The methods are compared using simple phantoms of line sources and water-filled circular and elliptical cylinders.