Narrow Results

Driver fatigue can be detected by constructing a discriminant mode using some features obtained from physiological signals. There exist two major challenges of this kind of methods. One is how to collect physiological signals from subjects while they are driving without any interruption. The other is to find features of physiological signals that are of corresponding change with the loss of attention caused by driver fatigue. Driving fatigue is detected based on the study of surface electromyography (EMG) and electrocardiograph (ECG) during the driving period. The noncontact data acquisition system was used to collect physiological signals from the biceps femoris of each subject to tackle the first challenge. Fast independent component analysis (FastICA) and digital filter were utilized to process the original signals. Based on the statistical analysis results given by Kolmogorov–Smirnov Z test, the peak factor of EMG (p < 0.001) and the maximum of the cross-relation curve of EMG and ECG (p < 0.001) were selected as the combined characteristic to detect fatigue of drivers. The discriminant criterion of fatigue was obtained from the training samples by using Mahalanobis distance, and then the average classification accuracy was given by 10-fold cross-validation. The results showed that the method proposed in this paper can give well performance in distinguishing the normal state and fatigue state. The noncontact, onboard vehicle drivers' fatigue detection system was developed to reduce fatigue-related risks.

The output power of wind turbine has great relation with its health state, and the health status assessment for wind turbines influences operational maintenance and economic benefit of wind farm. Aiming at the current problem that the health status for the whole machine in wind farm is hard to get accurately, in this paper, we propose a health status assessment method in order to assess and predict the health status of the whole wind turbine, which is based on the power prediction and Mahalanobis distance (MD). Firstly, on the basis of Bates theory, the scientific analysis for historical data from SCADA system in wind farm explains the relation between wind power and running states of wind turbines. Secondly, the active power prediction model is utilized to obtain the power forecasting value under the health status of wind turbines. And the difference between the forecasting value and actual value constructs the standard residual set which is seen as the benchmark of health status assessment for wind turbines. In the process of assessment, the test set residual is gained by network model. The MD is calculated by the test residual set and normal residual set and then normalized as the health status assessment value of wind turbines. This method innovatively constructs evaluation index which can reflect the electricity generating performance of wind turbines rapidly and precisely. So it effectively avoids the defect that the existing methods are generally and easily influenced by subjective consciousness. Finally, SCADA system data in one wind farm of Fujian province has been used to verify this method. The results indicate that this new method can make effective assessment for the health status variation trend of wind turbines and provide new means for fault warning of wind turbines.

The visualization of patterns related to chaos is a challenge for those who are part of today's dynamical systems community, especially when we consider the aim of providing users with the ability to visually analyze and explore large, complex datasets related to chaos. Thus visualization could be considered a useful element in the discovery of unexpected relationships and dependencies that may exist inside the domain of chaos, both in the phase and the parameter spaces. In the second part of "A Gallery of Chua attractors", we presented an overview of forms which can only be produced by the physical circuit. In Part III, we illustrate the variety and beauty of the strange attractors produced by the dimensionless version of the system. As in our earlier work, we have used ad hoc methods, such as bifurcation maps and software tools, allowing rapid exploration of parameter space. Applying these techniques, we show how it is possible, starting from attractors described in the literature, to find new families of patterns, with a special focus on the cognitive side of information seeking and on qualitative processes of change in chaos, thus demonstrating that traditional categories of chaos exploration need to be renewed.

After a brief introduction to dimensionless equations for Chua's oscillator, we show 150 attractors, which we represent using three-dimensional images, time series and FFT diagrams. For the most important patterns, we also report Lyapunov exponents. To show the position of dimensionless attractors in parameter space, we use parallel coordinate techniques that facilitate the visualization of high dimensional spaces. We use Principal Components Analysis (PCA) and Mahalanobis Distance to provide additional tools for the exploration and visualization of the structure of the parameter space.

When a control chart signals, it shows the process parameters have changed due to assignable cause(s). However, control chart signal is not the real time of a change in the process. Knowing the real time of change would simplify the detection and elimination of the assignable causes of variation. In this paper, a two-stage process is considered when the mean values of quality characteristics are changed under step shift and linear drift. First, a control chart based on the discriminant analysis (DA) is utilized to monitor the process. Then, when the out-of-control signal is received, the maximum likelihood estimator (MLE) based on the DA statistics, and clustering approach based on Mahalanobis distance of residuals are developed to estimate the real time of the change. The performances of the proposed estimators under different shifts are evaluated through numerical examples and a real case. The results indicate the better performance of the clustering approach rather than the MLE in most cases under both step shift and drift.

Current genotype-calling methods such as Robust Linear Model with Mahalanobis Distance Classifier (RLMM) and Corrected Robust Linear Model with Maximum Likelihood Classification (CRLMM) provide accurate calling results for Affymetrix Single Nucleotide Polymorphisms (SNP) chips. However, these methods are computationally expensive as they employ preprocess procedures, including chip data normalization and other sophisticated statistical techniques. In the small sample case the accuracy rate may drop significantly. We develop a new genotype calling method for Affymetrix 100 k and 500 k SNP chips. A two-stage classification scheme is proposed to obtain a fast genotype calling algorithm. The first stage uses unsupervised classification to quickly discriminate genotypes with high accuracy for the majority of the SNPs. And the second stage employs a supervised classification method to incorporate allele frequency information either from the HapMap data or from a self-training scheme. Confidence score is provided for every genotype call. The overall performance is shown to be comparable to that of CRLMM as verified by the known gold standard HapMap data and is superior in small sample cases. The new algorithm is computationally simple and standalone in the sense that a self-training scheme can be used without employing any other training data. A package implementing the calling algorithm is freely available at .

A front-end method based on random forest proximity distance (PD) is used to screen the test set to improve protein–protein interaction site (PPIS) prediction. The assessment of a distance metric is done under the assumption that a distance definition of higher quality leads to higher classification. On an independent test set, the numerical analysis based on statistical inference shows that the PD has the advantage over Mahalanobis and Cosine distance. Based on the fact that the proximity distance depends on the tree composition of the random forest model, an iterative method is designed to optimize the proximity distance, which adjusts the tree composition of the random forest model by adjusting the size of the training set. Two PD metrics, 75PD and 50PD, are obtained by the iterative method. On two independent test sets, compared with the PD produced by the original training set, the values of 75PD in Matthews correlation coefficient and F₁ score were higher, and the differences between them were statistically significant. All numerical experiments show that the closer the distance between the test data and the training data, the better the prediction results of the predictor. These indicate that the iterative method can optimize proximity distance definition and the distance information provided by PD can be used to indicate the reliability of prediction results.

In intelligent vehicle system, it is significant to detect and identify road markings for vehicles to follow traffic regulation. This paper proposes a method to recognize direction markings on road surface, which is on the basis of detected lanes and uses Hu moments. First of all, the detection of lanes is based on horizontal luminance difference, which converts the RGB color image to the luminance image, calculates the horizontal luminance difference, obtains the candidate points of lanes' edge and uses least square method to fit the lanes. Secondly, with the detected lines as guide for the search of candidate marking, the paper extracts Hu moments of candidate marking, calculates its Mahalanobis distance to every marking type and classifies it to the type which has the minimal distance with the candidate marking. From the simulation results, the method to detect lanes is more effective and time-efficient than canny or sobel edge detection methods; the method to recognize direction marking is effective and has a high accuracy.