The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
In this paper, we present a non-parametric dataanalytic soft-error detector. Our detector uses the key properties of Gaussian process regression. First, because Gaussian process regression provides confidence on the prediction, this confidence can be used to automatize construction of the detection range. Second, because the correlation model of a Gaussian process captures the similarity among neighboring...
We have proposed a concept for classification interesting points in images by means of a machine learning approach. The basic idea is that each interesting point detected in an image is classified either as a point belonging to some trained model (e.g. corner of a license plate) or not. During the first stage, we detected interesting points in a set of images by the well-known SURF method. Then we...
Traffic light detection (TLD) is a vital part of both intelligent vehicles and driving assistance systems (DAS). General for most TLDs is that they are evaluated on small and private datasets making it hard to determine the exact performance of a given method. In this paper we apply the state-of-the-art, real-time object detection system You Only Look Once, (YOLO) on the public LISA Traffic Light...
This paper discusses utilizing sparse autoencoders for building regression models in order to predict real-valued timeseries data. The focus of this research is on exploiting and learning from the determining features of continuous data via stacked autoencoders, thus increasing the prediction accuracy of regression method. Archi-tecture comprising different layers of sparse autoencod-ers, where each...
Multi-label learning is widely applied in many tasks, where an object possesses multiple concepts with each represented by a class label. Previous studies on multi-label learning have focused on a fixed set of class labels, i.e., the class label set of test data is the same as that in the training set. In many applications, however, the environment is open and new concepts may emerge with previously...
In order to meet the requirements of traffic data fusion for real-time urban traffic state estimation, a new kind of fusion structure model is proposed. This fusion model consists of both spatial fusion and temporal fusion. First we use the power average operator as spatial fusion method. Then we propose a temporal correlation based data compression (TCDC) algorithm, based on segment linear regression...
Short-term traffic flow forecasting has been a crucial component in the area of intelligent transportation systems (ITS), which plays a significant role in operating traffic management systems and dynamic traffic assignment effectively as well as proactively. In this paper, a novel short-term traffic flow prediction method called Ensemble Real-time Sequential Extreme Learning Machine (ERS-ELM) with...
Recent high profile security breaches have highlighted the importance of insider threat detection systems for cybersecurity. However, issues such as high false positive rates and concerns over data privacy make it difficult to predict performance within an enterprise environment. These and other issues limit an organization's ability to effectively apply these tools. In this paper, we present an approach...
Travel time parameters obtained from road traffic sensors data play an important role in traffic management practice. In this paper, a travel time analysis and prediction model was established for urban road traffic sensors data based on the change point analysis algorithm and ARIMA model. Firstly, time series of travel time parameters were clustered by using change point mining algorithm after traffic...
The prediction of both, vehicular traffic and communication connectivity are important research topics. In this paper, we propose the usage of innovative machine learning approaches for these objectives. For this purpose, Poisson Dependency Networks (PDNs) are introduced to enhance the prediction quality of vehicular traffic flows. The machine learning model is fitted based on empirical vehicular...
A detection methodology for marine debris presence after a natural disaster is described. The methodology is based both on a predictive model and a Bayesian hierarchical spatial method. The chosen fusion approach relies on auto-logistic regression to weight the outputs of multiple target detection algorithms, as well as to capture the intrinsic processes related to the presence of marine debris. The...
Anomaly detection in streaming data is of high interest in numerous application domains. In this paper, we propose a novel one-class semi-supervised algorithm to detect anomalies in streaming data. Underlying the algorithm is a fast and accurate density estimator implemented by multiple fully randomized space trees (RS-Trees), named RS-Forest. The piecewise constant density estimate of each RS-tree...
The online anomaly detection has been propounded as the the key idea of monitoring fault of large-scale sensor nodes in Internet of Things. Although the exciting progresses of research have been made in online anomaly detection area, the highly dynamic distribution makes the anomaly detection scheme difficult to be used in online manner. This paper presents an online anomaly learning and forecasting...
In this study, we propose a novel solution to regulate the amount of interest points extracted from an image without significant additional computational cost. Our method acts at the very beginning of the detection process by using a corner occurrence model in order to predict the optimal threshold for a user-defined number of detections. Compared to existing approaches which guarantee a reasonable...
Architectural tactics such as heartbeat, resource pooling, and scheduling, offer proven solutions for systematically increasing the reliability, security, performance, and other critical characteristics of a software system. Current literature on architectural tactics advocates a requirements-driven approach for deciding when and where tactics should be used in order to address specific quality concerns...
Travel time information is a fundamental component in Advanced Traveler Information System. In this paper, we propose a short-term travel time estimation and prediction framework for long freeway corridor, considering measurements from vehicle detectors (VD) and floating car data (FCD). The modeling approach is based on a modified Nearest-Neighborhood (NN) model with threshold and a regression model...
Faced with a deluge of data, an analyst must ask ``what data records are important?'' This paper answers that question by first defining a continuous spectrum of data record significance: ``known'', ``anomalous'', ``interesting'', ``novel'', and ``noise''. The definition has a geometric interpretation in that the significance of a data record in a predictor system is inversely proportional to it's...
The production of light ions such as protons, neutrons, deuterons, tritons, 3He and 4He from heavy ion interactions still remains as a key issue to be investigated for the purposes of radiation protection in space. Ultimately, we will produce double-differential spectra for high-energy light ion production from several heavy-ion experiments conducted at the Heavy-Ion Medical Accelerator in Chiba (HIMAC)...
Soft sensors are a valuable alternative to the traditional hardware sensors, which are indispensable in configuration of modern systems. They are often used in process industry, but other applications are possible. This paper describes a possible application of soft sensor for faulty measurement detection and reconstruction in urban traffic. One of the key indicators of traffic control quality in...
This paper provides a neural network model to address the problem of travel time prediction. A single segment model based on the state space neural network is used for modeling traffic flow on one single signalized segment. Thus, modelling a longer arterial covering several controlled intersections is conducted by assembling each individual segment models. This reduces significantly the amount of...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.