Unsupervised event detection and classification of multichannel signals

In this paper, we present a new unsupervised method to classify a set of Multichanel Signals (MC) with unknown events. Each signal is characterized by a sequence of events where the number of events, the start time and the duration between events can change randomly. The proposed method helps in the classification and event detection of the MC signals by an expert which usually becomes a tedious and difficult task. To this end, first, the problem of classification of MC signals characterized by a succession of events is analyzed by transforming the MC signals into a set of temporal sequences of easy interpretation. The algorithm detects events by means of an optimal unsupervised classification. It is not necessary to know the nature of the events and formulate hypotheses regarding their behavior. Then, a set of multichannel electromyographic (EMG) signals with events is generated. These MC signals are used to test the proposed method.     

Clump splitting via bottleneck detection and shape classification

Under-segmentation of an image with multiple objects is a common problem in image segmentation algorithms. This paper presents a novel approach for splitting clumps formed by multiple objects due to under-segmentation. The proposed algorithm includes three steps: (1) decide whether to split a candidate connected component by application-specific shape classification; (2) find a pair of points for clump splitting and (3) join the pair of selected points. In the first step, a shape classifier is applied to determine whether a connected component should be split. In the second step, a pair of points for splitting is detected using a bottleneck rule, under the assumption that the desired objects have roughly a convex shape. In the third step, the selected splitting points from step two are joined by finding the optimal splitting line between them, based on minimizing an image energy. The shape classifier is built offline via various shape features and a support vector machine. Steps two and three are application-independent. The performance of this method is evaluated using images from various applications. Experimental results show that the proposed approach outperforms the state-of-the-art algorithms for the clump splitting problem.     

Minimum volume ellipsoid classification model for contamination event detection in water distribution systems

The presented study features an event detection model alerting for contamination events in water distribution systems. The developed model comprises a minimum volume ellipsoid (MVE) classifier, detecting outlier measurements, and a following sequence analysis utilizing the MVE binary output, for the classification of events. The model is updated continuously and exploits a constantly growing data base. The MVE enables simultaneous analysis of the water quality parameters. The multivariate analysis explores the relations between water quality parameters and detects changes in their common patterns. The suggested model applied an un-supervised classification method, eliminates the need for simulated events examples in the classifier construction. In the absent of satisfying information regarding the influence of contamination event on the parameter measurements, eliminating the use of any assumption contributes to the model reliability and generality. The model was trained on a real water utility data, and tested on randomly simulated events that were superimposed on the original data base. The model showed high accuracy and detection ability compared to previous studies.     

On integrating event definition and event detection

We develop, in this paper, a representation of time and events that supports a range of reasoning tasks such as monitoring and detection of event patterns which may facilitate the explanation of root cause(s) of faults. We shall compare two approaches to event definition: the active database approach in which events are defined in terms of the conditions for their detection at an instant, and the knowledge representation approach in which events are defined in terms of the conditions for their occurrence over an interval. We shall show the shortcomings of the former definition and employ a three-valued temporal first order nonmonotonic logic, extended with events, in order to integrate both definitions.