Search results for "Mach"

Head Pose Estimation for Sign Language Video

2013

We address the problem of estimating three head pose angles in sign language video using the Pointing04 data set as training data. The proposed model employs facial landmark points and Support Vector Regression learned from the training set to identify yaw and pitch angles independently. A simple geometric approach is used for the roll angle. As a novel development, we propose to use the detected skin tone areas within the face bounding box as additional features for head pose estimation. The accuracy level of the estimators we obtain compares favorably with published results on the same data, but the smaller number of pose angles in our setup may explain some of the observed advantage.

Assessment of qualitative judgements for conditional events in expert systems

1991

Computer Simulation for the Study of CNC Feed Drives Dynamic Behavior and Accuracy

2007

In the application of CNC feed drives it is desirable to predict the servo performance. By using computer simulation techniques it is possible to construct an accurate model of the servo drive. This simulation procedure makes it possible to anticipate machine design problems and correct them. This paper deals with a model of a feed drive, which consists of a motion control system driven by a DC motor. Both position and velocity feedback loops are present in the structure of the system. By means of MATLAB & Simulink software, simulation diagrams were built in order to test the behaviour of the system. Experimental data are also presented in order to confirm the accuracy of the theoretical mo…

Towards a SDN-based architecture for analyzing network traffic in cloud computing infrastructures

2015

Currently, network traffic monitoring tools do not fit well in the monitoring of cloud computing infrastructures. These tools are not integrated with the control plane of the cloud computing stack. This lack of integration causes a deficiency in the handling of the re-usage of IP addresses along virtual machines, a lack of adaption and reaction on highly frequent topology changes, and a lack of accuracy in the metrics gathered for the networking traffic flowing along the cloud infrastructure. The main contribution of this paper is to provide a novel SDN-based architecture to carry out the monitoring of network traffic in cloud infrastructures. The architecture in based on the integration be…

Review of web-based information security threats in smart grid

2017

The penetration of digital devices in Smart Grid has created a big security issue. OWASP is an online community of security professionals that identifies the most critical web application security risk in IT domain. Smart Grid also uses client-server based web-applications to collect and disseminate information. Therefore, Smart Grid network is analogous to IT network and similar kind of risk exists in the Smart Grid. This paper review the security risk in Smart Grid domain with reference to OWASP study. The Smart Grid security is more biased towards vulnerabilities associated with a machine to machine communication. Methodology to minimise the risk of attack is also discussed in this resea…

Convolutional Long Short-Term Memory Network for Multitemporal Cloud Detection Over Landmarks

2019

In this work, we propose to exploit both the temporal and spatial correlations in Earth observation satellite images through deep learning methods. In particular, the combination of a U-Net convolutional neural network together with a convolutional long short-term memory (LSTM) layer is proposed. This model is applied for cloud detection on MSG/SEVIRI image time series over selected landmarks. Implementation details are provided and our proposal is compared against a standard SVM and a U-Net without the convolutional LSTM layer but including temporal information too. Experimental results show that this combination of networks exploits both the spatial and temporal dependence and provides st…

Fallzahlplanung in referenzkontrollierten Diagnosestudien

2002

Purpose: A tutorial illustration of a flexible approach to determine the sample size in reference-controlled diagnostic trials. Materials and Methods: Assuming the usual setting of a new diagnostic method to be compared with a reference method, the emphasis is on the sensitivity of the new method in comparison with the reference method, using a binary outcome (positive versus negative) for both methods. Based on the confidence interval of the sensitivity, a simple but flexible procedure for determining the sample size is described, which incorporates clinically interpretable information. The procedure is illustrated by the fictious planning of a trial to assess the diagnostic value of MRI v…

Psychological Influence of Double-Bind Situations in Human-Agent Interaction

2007

This paper presents a new approach to integrate artificial intelligence in virtual environments. The system presented deals in a separated way the visualization and intelligence modules, applying in this last case a distributed approach (multi-agent systems) so that scalable applications may be built. Therefore, it is necessary to define agent architectures that allow agents to be integrated in the VW. Thus, a designer is abstracted from the peculiarities of interacting with a virtual environment. There is a first prototype of the framework using JADE as the supporting multi-agent systems platform.

Correlation-Based and Contextual Merit-Based Ensemble Feature Selection

2001

Recent research has proved the benefits of using an ensemble of diverse and accurate base classifiers for classification problems. In this paper the focus is on producing diverse ensembles with the aid of three feature selection heuristics based on two approaches: correlation and contextual merit -based ones. We have developed an algorithm and experimented with it to evaluate and compare the three feature selection heuristics on ten data sets from UCI Repository. On average, simple correlation-based ensemble has the superiority in accuracy. The contextual merit -based heuristics seem to include too many features in the initial ensembles and iterations were most successful with it.

Extracting information from support vector machines for pattern-based classification

2014

Statistical machine learning algorithms building on patterns found by pattern mining algorithms have to cope with large solution sets and thus the high dimensionality of the feature space. Vice versa, pattern mining algorithms are frequently applied to irrelevant instances, thus causing noise in the output. Solution sets of pattern mining algorithms also typically grow with increasing input datasets. The paper proposes an approach to overcome these limitations. The approach extracts information from trained support vector machines, in particular their support vectors and their relevance according to their coefficients. It uses the support vectors along with their coefficients as input to pa…