Local Event Detection Research Articles

Distributed polymer optical fiber sensors using digital I-OFDR for geotechnical infrastructure health monitoring

We present a distributed polymer optical fiber sensor system for deformation monitoring of geotechnical infrastructure. The sensor system is based on the digital incoherent optical frequency domain reflectometry (I-OFDR) for the detection of local strain events along a perfluorinated polymer optical fiber (PF-POF) used as a sensing fiber. For the best possible load transfer, the PF-POFs were integrated onto geotextiles which pose a sensor carrier for the sensing fiber. By using elastic PF-POF instead of a standard glass fiber as a sensing fiber the strain range of geotextile-integrated fiber optic sensors could be extended up to 10 % in accordance with the end-user requirements. In order to reduce overall costs of the sensor system and achieve its robustness required for use on site, the standard vector network analyzer used by default was replaced by an integrated digital board for optical frequency domain analysis. This implementation showed a significant improvement in the dynamic range and sensitivity compared to the previous measurement system based on the conventional network analyzer. The functionality of the entire sensor system was proven in the field test during the installation of two geomats with integrated PF-POFs embedded into a road embankment in a section of the B 91 federal road near Werschen in the greater Leipzig area. To monitor the installation-related changes in the condition of both sensor-based geomats several control measurements on the embedded PF-POFs and co-integrated reference glass fibers were carried as the construction works progressed. The measurements recorded in the field confirm the great potential of the selected PF-POF used as the distributed fiber optic strain sensor. The future management of the remaining challenges of industrial strain-coupled sensor integration can result in a successful market launch of the PF-POF-sensor integrated into geosynthetics. The sensor system is ready to be used for practical application.

Interaction graph, topical communities, and efficient local event detection from social streams

Social networks have become an essential part of daily life, and hence every real-world activity finds its place in this virtual world. The present paper proposes a methodology to find localized micro-events from the social network stream. The method is named CommunityINDICATOR. A concept of ‘separation of concerns’ from the software design principle is incorporated in the methodology to reduce the execution time drastically from existing state-of-the-art methods of event detection. In order to reduce the execution time, the algorithm first generates an interaction graph from the social stream and applies community detection followed by a clustering algorithm onto it to detect micro-level events. Experiments have been conducted on Twitter data stream of 5 different cities on three different continents with the size of 2 million tweets. We have used well known quality metrics such as precision, recall, F1-score, accuracy, and execution time to compare performance with other state-of-the-art methodologies. The proposed CommunityINDICATOR provides up to 30% higher accuracy than EvenTweet and SEDTWik in similar execution times. An improvement of 11% to 51% and 17% to 57% in execution time is observed for the proposed algorithm in comparison with TwiiterNews and EventX, respectively, for different datasets.

WINTENDED: WINdowed TENsor decomposition for Densification Event Detection in time-evolving networks

Densification events in time-evolving networks refer to instants in which the network density, that is, the number of edges, is substantially larger than in the remaining. These events can occur at a global level, involving the majority of the nodes in the network, or at a local level involving only a subset of nodes.While global densification events affect the overall structure of the network, the same does not hold in local densification events, which may remain undetectable by the existing detection methods. In order to address this issue, we propose WINdowed TENsor decomposition for Densification Event Detection (WINTENDED) for the detection and characterization of both global and local densification events. Our method combines a sliding window decomposition with statistical tools to capture the local dynamics of the network and automatically find the irregular behaviours. According to our experimental evaluation, WINTENDED is able to spot global densification events at least as accurately as its competitors, while also being able to find local densification events, on the contrary to its competitors.

Local Event Detection Scheme by Analyzing Relevant Documents in Social Networks

In this paper, we propose a local event detection scheme by analyzing relevant documents in social networks to improve the accuracy of event detection. To detect local events by using geographical data, the proposed scheme embeds them using a geographical data dictionary and generates a weighted keyword graph using social network characteristics. The data left by users in social networks include not only postings but also related documents such as comments and threads. In this way, the proposed scheme detects a local event based on a keyword graph that is constructed through the analysis of the relevant documents. This can improve the accuracy of local event detection by analyzing relevant documents embedded with region-related information using a geographical data dictionary, without requiring users to tag geographic data. In order to verify the superiority of the proposed scheme, we compare it with the existing event detection schemes through various performance evaluations.

Local earthquakes detection: A benchmark dataset of 3-component seismograms built on a global scale

Machine learning is becoming increasingly important in scientific and technological progress, due to its ability to create models that describe complex data and generalize well. The wealth of publicly-available seismic data nowadays requires automated, fast, and reliable tools to carry out a multitude of tasks, such as the detection of small, local earthquakes in areas characterized by sparsity of receivers. A similar application of machine learning, however, should be built on a large amount of labeled seismograms, which is neither immediate to obtain nor to compile. In this study we present a large dataset of seismograms recorded along the vertical, north, and east components of 1487 broad-band or very broad-band receivers distributed worldwide; this includes 629,095 3-component seismograms generated by 304,878 local earthquakes and labeled as EQ, and 615,847 ones labeled as noise (AN). Application of machine learning to this dataset shows that a simple Convolutional Neural Network of 67,939 parameters allows discriminating between earthquakes and noise single-station recordings, even if applied in regions not represented in the training set. Achieving an accuracy of 96.7, 95.3, and 93.2% on training, validation, and test set, respectively, we prove that the large variety of geological and tectonic settings covered by our data supports the generalization capabilities of the algorithm, and makes it applicable to real-time detection of local events. We make the database publicly available, intending to provide the seismological and broader scientific community with a benchmark for time-series to be used as a testing ground in signal processing.

Collective Memory for Detecting Nonconcurrent Clones: A Localized Approach for Global Topology and Identity Tracing in IoT Networks

Clone attack is considered as a severely destructive threat in Internet of Things (IoT), because: 1) the attack may be easily launched due to the deficiency of hardware architecture and the limited resources against physical capture and compromise and 2) it may trigger a large variety of insider and outsider attacks. Different from traditional clone attack detection approaches that ground on a large amount of data traversing the network (e.g., locations and identities), this article tackles this problem by answering the following fundamental questions: do we really need so much raw information? whether there is an alternative for local event detection by a node far away from that event? when acquiring/tracing global knowledge of a system/network, do we really need a global collection effort? These questions are of much importance in a large variety of networks. Specifically, this article provides a collective memory design for global topology and identity tracing (GTI Tracing), via a localized computing paradigm within neighborhood. This localized paradigm computationally builds a connection of identity and topology from time and space domain to a new computation domain. Such a computation domain retains four properties: 1) transitivity; 2) global convergence; 3) determinacy; and 4) causality. With byte-size information at an arbitrary device, it can recover and keep tracing global topology and identity information, and thus providing deterministic detection of clones. Both theoretical analysis and experimental study have shown the advantages of the proposed design in both detection accuracy and privacy protection, at a cost of light communication, storage, and computation overhead at each device.

Abnormal event detection in surveillance videos based on low-rank and compact coefficient dictionary learning

In this paper, a novel approach to abnormal event detection in crowded scenes is presented based on a new low-rank and compact coefficient dictionary learning (LRCCDL) algorithm. First, based on the background subtraction and binarization of surveillance videos, we construct a feature space by extracting the histogram of maximal optical flow projection (HMOFP) feature of the foreground from a normal training frame set. Second, in the training stage, a new joint optimization of the nuclear-norm and l2, 1-norm is applied to obtain a compact coefficient low-rank dictionary. Third, in the detection stage, l2, 1-norm optimization is utilized to obtain the reconstruction coefficient vectors of the testing samples. Note that the l2, 1-norm forces the reconstruction vectors of all the testing samples to compactly surround the same center in the training stage, such that the reconstruction errors of abnormal testing samples are different from those of normal ones. Finally, a reconstruction cost (RC) is introduced to detect abnormal frames. Experimental results on both global and local abnormal event detection show the effectiveness of our algorithm. Based on comparisons with state-of-the-art methods employing various criteria, the proposed algorithm achieves comparable detection results.

AED-Net: An Abnormal Event Detection Network

Abstract It has long been a challenging task to detect an anomaly in a crowded scene. In this paper, a self-supervised framework called the abnormal event detection network (AED-Net), which is composed of a principal component analysis network (PCAnet) and kernel principal component analysis (kPCA), is proposed to address this problem. Using surveillance video sequences of different scenes as raw data, the PCAnet is trained to extract high-level semantics of the crowd’s situation. Next, kPCA, a one-class classifier, is trained to identify anomalies within the scene. In contrast to some prevailing deep learning methods, this framework is completely self-supervised because it utilizes only video sequences of a normal situation. Experiments in global and local abnormal event detection are carried out on Monitoring Human Activity dataset from University of Minnesota (UMN dataset) and Anomaly Detection dataset from University of California, San Diego (UCSD dataset), and competitive results that yield a better equal error rate (EER) and area under curve (AUC) than other state-of-the-art methods are observed. Furthermore, by adding a local response normalization (LRN) layer, we propose an improvement to the original AED-Net. The results demonstrate that this proposed version performs better by promoting the framework’s generalization capacity.

Unsupervised deep learning system for local anomaly event detection in crowded scenes

Anomaly detection in video surveillance is a significant research subject because of its immense use in real-time applications. These days, open spots like hospitals, traffic areas, airports are monitored by video surveillance cameras. Strange occasions in these recordings have alluded to the anomaly. Unsupervised anomaly detection in the video be endowed with many challenges as there is no exact definition of abnormal events. It varies as for various situations. This paper aims to propose an effective unsupervised deep learning framework for video anomaly detection. Raw image sequences are combined with edge image sequences and given as input to the convolutional auto encoder-ConvLSTM model. Experimental evaluation of the proposed work is performed in three different benchmark datasets such as Avenue, UCSD ped1 and UCSD ped2. The proposed method Hybrid Deep Learning framework for Video Anomaly Detection (HDLVAD) reaches better accuracy compared to existing methods. Investigating video streaming in big data is our further research work.

User-driven geolocated event detection in social media

Event detection is one of the most important research topics in social media analysis. Despite this interest, few researchers have addressed the problem of identifying geolocated events in an unsupervised way, and none includes user interests during the process. In this paper, we tackle the problem of local event detection from social media data. We present a method to automatically identify events by evaluating the burstiness of hashtags in a geographical area and a time interval, and at the same time integrating user feedback. We devise two algorithms to discover user-driven events. The first one relies on an exact enumeration process, while the other directly samples the space of events. In our empirical study, we provide evidence that geolocated events cannot be detected by non location-aware methods. We also show that our methods (i) outperform by a factor of two to several orders of magnitude state-of-the-art methods designed to discover geolocated events, (ii) are more robust to noise, and (iii) produce high quality events with respect to user interests.

Event Detection and Recommendation Using Social Media for Hyperlocal Communities

Individuals often rely on newspapers and television to gather information about the communities they live in to be aware of and to engage with their communities. Social media have become increasingly important as a new source of local information. Thus far, the literature on using social media to create desirable societal benefits, such as in civic awareness and engagement, is still in its infancy in the Information System (IS) field. One key challenge in this research stream is to timely and accurately distill local information from noisy Twitter streams to community members. In this work, we develop several machine learning techniques and a mobile application utilizing these techniques to facilitate information seeking of local communities on Twitter. Specifically, we first propose a novel local event detection algorithm which scans spatially and temporally to detect local events as unusual spikes from a large-scale Twitter feeds. Our algorithm significantly improves several competing methods with much higher precision and recall in detecting local events on a real-world dataset of millions of geotagged tweets. Next, the volume of detected local events can be enormous, especially in large and populous local areas. This motivates us to develop a personalized local event recommender system that systematically integrates several contextual cues such as topical, geographical and social proximity in a learning to rank framework. Based on a comprehensive evaluation with two large-scale datasets of local events, the results show that our approach significantly outperforms competitive methods in precision as well as its robustness. Finally, we build a mobile application to demonstrate the practical value of our event detection and recommendation techniques. This application automatically infers and recommends neighborhood-specific information based on Twitter feeds. We test its effectiveness and usefulness through a comprehensive user study. The outcomes demonstrate that most participants prefer it over Twitter as it is easier to use for exploring their neighborhoods and provides better relevant information about their neighborhoods.

Time-of-Flight Neutron Imaging for the Distinction of Super-Cooled Water and Ice during PEFC Cold-Starts

The start-up of polymer electrolyte fuel cells (PEFCs) at sub-zero temperatures (cold-start) is a challenge for fuel cell designers and operators. The freezing of water in the cell can cause shutdown of the cell and severe damages to fuel cell components. However, water produced in the electrochemical reaction is pure enough to be present in the super-cooled state (liquid) at sub-zero temperatures (1). Methods that allow visualizing the location of freezing events during cold-starts help to understand which parameters influence the phase transitions. For this purpose our group has been using energy selective neutron imaging methods since 2013 (2). We exploit the fact that the attenuation for ice and super-cooled water differs at low neutron energies (wavelength > 4 Å) while it is identical for high energies (3). When water freezes in the pores of the catalyst and/or gas diffusion layer, its volume expands in all directions and the density decreases. To account for the density change, we record images in two different energy windows - one corresponding to high and the other to low energy neutrons – and obtain a relative attenuation image, which allows us to determine the state of aggregation. Energy selective neutron imaging with a neutron filter was successfully applied while operating a fuel cell with an active area of 50 cm2 for the detection of local freezing events (4). However, the obtained contrast between liquid water and ice was very low (1.6%), making the method particularly sensitive to biases. To increase the contrast, we use time-of-flight neutron imaging (TOF-NI), a method based on neutron pulses (either at a pulsed neutron source or with the help of a rotating chopper disk) where the neutron energy is determined by the time lag between the pulse and the moment the neutron reaches the detector. Compared to conventional TOF-NI, which uses short neutron pulses (duty cycles smaller than 1%), we use a chopper disk with broad slits making a duty cycle of approximately 30% and obtain a much higher neutron flux at the cost of lower energy resolution. Here, we will present experimental results obtained during isothermal cold-starts with a cell having an active area of 4.4 cm2. Simultaneously to neutron imaging, we measured the heat fluxes in and out of the cell with a setup described in other references (5). Figure 1 contains an example of the results obtained during a cold-start performed at -7.5°C. Figure 1 (a) depicts the evolution of voltage and heat flux over time during the cold-start while Fig. 1 (b) shows the relative attenuation (ratio between low and high energy attenuation) averaged over the whole active area (A), and over the channels B and C (the different areas are represented on Fig. 1 (e)). During phase I, larger accumulations of super-cooled water were present in the gas flow channels. The phase transition of super-cooled water to ice occurred 62 minutes after starting to draw current. This is visible from the voltage drop and the peak in heat flux due to the latent heat that is released during freezing of water [Fig 1 (a)]. Phase II corresponds to the time after the freezing event where the cell was not operated and its temperature was kept constant at -7.5°C. Figure 1 (c) and (d) depict the high energy attenuation images, from which the water thickness is retrieved by using the Beer-Lambert law. Those images are representative for classical transmission images without energy selectivity. The images representing the relative attenuation (attenuation at low energy divided by attenuation at high energy) are shown in Figure 1 (e) and (f). For clarity, only channel water is shown in these images. By comparing the classical attenuation based images [Fig 1 (c-d)] with the energy selective approach [Fig 1 (e-f)] it becomes evident that the high energy attenuation does not clearly differ for phase I and II. In contrast, the relative attenuation for ice measured in phase II [Fig 1 (b and f)] is – as expected – lower compared to the one measured for super-cooled water in phase I [Fig 1 (b and e)]. By using TOF-NI we could improve the contrast between super-cooled water and ice to reach approximately 5%. P. Oberholzer et. al, J. Electrochem. Soc., 159, B235 (2011). J. Biesdorf et. al, Phys. Rev. Lett., 112 (2014). L. Torres et. al, Nuc.l Instrum. Meth. B, 251, 304 (2006). P. Stahl et. al, J. Electrochem. Soc., 163, F1535 (2016). M. Cochet et. al, Submitted for publication (2018). Figure 1

Similarity based image selection with frame rate adaptation and local event detection in wireless video sensor networks

Wireless Video Sensor Networks (WVSNs7unding environmental information. Those sensor nodes can locally process the information and then wirelessly transmit it to the coordinator and to the sink to be further processed. As a consequence, more abundant video and image data are collected. In such densely deployed networks, the problem of data redundancy arises when information are gathered from neighboring nodes. To overcome this problem, one important enabling technology for WVSN is data aggregation, which is essential to be cost-efficient. In this paper, we propose a new approach for data aggregation in WVSN based on images and shot similarity functions. It is deployed on two levels: the video-sensor node level and the coordinator level. At the sensor node level the proposed algorithms aim at reducing the number of frames sensed by the sensor nodes and sent to the coordinator. At the coordinator level, after receiving shots from different neighbouring sensor nodes, the similarity between these shots is computed to eliminate redundancies and to only send the frames which meet a certain condition to the sink. The similarity between shots is evaluated based on their color, edge and motion information. We evaluate our approach on a live scenario and compare the results with another approach from the literature in terms of data reduction and energy consumption. The results show that the two approaches have a significant data reduction to reduce the energy consumption, thus our approach tends to overcome the other one in terms of reducing the energy consumption related to the sensing process, and to the transmitting process while guaranteeing the detection of all the critical events at the node and the coordinator levels.

Abnormal global and local event detection in compressive sensing domain

Abnormal event detection, also known as anomaly detection, is one challenging task in security video surveillance. It is important to develop effective and robust movement representation models for global and local abnormal event detection to fight against factors such as occlusion and illumination change. In this paper, a new algorithm is proposed. It can locate the abnormal events on one frame, and detect the global abnormal frame. The proposed algorithm employs a sparse measurement matrix designed to represent the movement feature based on optical flow efficiently. Then, the abnormal detection mission is constructed as a one-class classification task via merely learning from the training normal samples. Experiments demonstrate that our algorithm performs well on the benchmark abnormal detection datasets against state-of-the-art methods.

GeoBurst+

The real-time discovery of local events (e.g., protests, disasters) has been widely recognized as a fundamental socioeconomic task. Recent studies have demonstrated that the geo-tagged tweet stream serves as an unprecedentedly valuable source for local event detection. Nevertheless, how to effectively extract local events from massive geo-tagged tweet streams in real time remains challenging. To bridge the gap, we propose a method for effective and real-time local event detection from geo-tagged tweet streams. Our method, named G eo B urst+ , first leverages a novel cross-modal authority measure to identify several pivots in the query window. Such pivots reveal different geo-topical activities and naturally attract similar tweets to form candidate events. G eo B urst+ further summarizes the continuous stream and compares the candidates against the historical summaries to pinpoint truly interesting local events. Better still, as the query window shifts, G eo B urst+ is capable of updating the event list with little time cost, thus achieving continuous monitoring of the stream. We used crowdsourcing to evaluate G eo B urst+ on two million-scale datasets and found it significantly more effective than existing methods while being orders of magnitude faster.

Abnormal event detection via covariance matrix for optical flow based feature

Abnormal event detection is one of the most important objectives in security surveillance for public scenes. In this paper, a new high-performance algorithm based on spatio-temporal motion information is proposed to detect global abnormal events from the video stream as well as the local abnormal event. We firstly propose a feature descriptor to represent the movement by adopting the covariance matrix coding optical flow and the corresponding partial derivatives of multiple connective frames or the patches of the frames. The covariance matrix of multi-RoI (region of interest) which consists of frames or patches can represent the movement in high accuracy. For public surveillance video, the normal samples are abundant while there are few abnormal samples. Thus the one-class classification method is suitable for handling this problem inherently. The nonlinear one-class support vector machine based on a proposed kernel for Lie group element is applied to detect abnormal events by merely training the normal samples. The computational complexity and time performance of the proposed method is analyzed. The PETS, UMN and UCSD benchmark datasets are employed to verify the advantages of the proposed method for both global abnormal and local abnormal event detection. This method can be used for event detection for a surveillance video and outperforms the state-of-the-art algorithms. Thus it can be adopted to detect the abnormal event in the monitoring video.

Local Seismic Events in the Area of Poland Based on Data from the PASSEQ 2006–2008 Experiment

PASSEQ 2006–2008 (Passive Seismic Experiment in TESZ; Wilde-Piorko et al. 2008) was the biggest passive seismic experiment carried out so far in the area of Central Europe (Poland, Germany, the Czech Republic and Lithuania). 196 seismic stations (including 49 broadband seismometers) worked simultaneously for over two years. During the experiment, multiple types of data recorders and seismometers were used, making the analysis more complex and time consuming. The dataset was unified and repaired to start the detection of local seismic events. Two different approaches for detection were applied for stations located in Poland. The first one used standard STA/LTA triggers (Carl Johnson’s STA/LTA algorithm) and grid search to classify and locate the events. The result was manually verified. The second approach used Real Time Recurrent Network (RTRN) detection (Wiszniowski et al. 2014). Both methods gave similar results, showing four previously unknown seismic events located in the Gulf of Gdansk area, situated in the southern Baltic Sea. In this paper we discuss both detection methods with their pros and cons (accuracy, efficiency, manual work required, scalability). We also show details of all detected and previously unknown events in the discussed area.

Single Layer Recurrent Neural Network for detection of swarm-like earthquakes in W-Bohemia/Vogtland—the method

In this paper, we present a new method of local event detection of swarm-like earthquakes based on neural networks. The proposed algorithm uses unique neural network architecture. It combines features used in other neural network concepts such as the Real Time Recurrent Network and Nonlinear Autoregressive Neural Network to achieve good performance of detection. We use the recurrence combined with various delays applied to recurrent inputs so the network remembers history of many samples. This method has been tested on data from a local seismic network in West Bohemia with promising results. We found that phases not picked in training data diminish the detection capability of the neural network and proper preparation of training data is therefore fundamental. To train the network we define a parameter called the learning importance weight of events and show that it affects the number of acceptable solutions achieved by many trials of the Back Propagation Through Time algorithm. We also compare the individual training of stations with training all of them simultaneously, and we conclude that results of joint training are better for some stations than training only one station.

A modular system for global and local abnormal event detection and categorization in videos

This paper presents a modular system for both abnormal event detection and categorization in videos. Complementary normalcy models are built both globally at the image level and locally within pixels blocks. Three features are analyzed: (1) spatio-temporal evolution of binary motion where foreground pixels are detected using an enhanced background subtraction method that keeps track of temporarily static pixels; (2) optical flow, using a robust pyramidal KLT technique; and (3) motion temporal derivatives. At the local level, a normalcy MOG model is built for each block and for each flow feature and is made more compact using PCA. Then, the activity is analyzed qualitatively using a set of compact hybrid histograms embedding both optical flow orientation (or temporal gradient orientation) and foreground statistics. A compact binary signature of maximal size 13 bits is extracted from these different features for event characterization. The performance of the system is illustrated on different datasets of videos recorded on static cameras. The experiments show that the anomalies are well detected even if the method is not dedicated to one of the addressed scenarios.

Network hydraulics inclusion in water quality event detection using multiple sensor stations data

Event detection is one of the current most challenging topics in water distribution systems analysis: how regular on-line hydraulic (e.g., pressure, flow) and water quality (e.g., pH, residual chlorine, turbidity) measurements at different network locations can be efficiently utilized to detect water quality contamination events. This study describes an integrated event detection model which combines multiple sensor stations data with network hydraulics. To date event detection modelling is likely limited to single sensor station location and dataset. Single sensor station models are detached from network hydraulics insights and as a result might be significantly exposed to false positive alarms. This work is aimed at decreasing this limitation through integrating local and spatial hydraulic data understanding into an event detection model. The spatial analysis complements the local event detection effort through discovering events with lower signatures by exploring the sensors mutual hydraulic influences. The unique contribution of this study is in incorporating hydraulic simulation information into the overall event detection process of spatially distributed sensors. The methodology is demonstrated on two example applications using base runs and sensitivity analyses. Results show a clear advantage of the suggested model over single-sensor event detection schemes.