Change Detector Research Articles

Enhanced Unsupervised Change Detector for Industrial Surveillance Systems

Detecting an intruder that is trespassing a prohibited area is a critical task of intelligent surveillance systems. This task requires a change detector to segment an intruder (foreground object) from the background. The task suffers the inherent drawbacks of change detectors due to the dual-camera sensor (color/IR), illumination changes, night time, static, and camouflaged foreground objects. This article proposes an enhanced unsupervised change detector to compensate for the aforementioned challenges for industrial sterile zone monitoring. The camera switch detection based on skewness patterns detects a switch between the dual-camera sensors (color/IR). The optimal color space selection based on the mean squared error will select tolerant color space (RGB/YCbCr) to illumination changes for modeling the background. In addition, the IR camera frames are contrast-enhanced to tackle the camouflaged intruders during the night. The incoming frames are split into respective channels before modeling the background. The background is modeled by Gaussian mixture models. The adaptive background model update scheme is proposed to tackle the various challenges posed by environment and object such as a static foreground object. The comparison is performed on three databases with top-ranked unsupervised change detection algorithms.

Improved Change Detector Using Dual-Camera Sensors for Intelligent Surveillance Systems

Unauthorized entrance in a prohibited area might create a security risk. An intelligent surveillance system should be able to mitigate such a problem by incorporating a sterile zone monitoring algorithm. The algorithm is challenged by a dual-camera sensors (color/IR), dynamic backgrounds, illumination changes, camouflaged, and static foreground objects, etc. This paper proposes an improved change detector (ICD) to mitigate the above-mentioned challenges. It employs a novel statistical decision criterion (SDC) based on skewness patterns. The SDC helps to differentiate time of day using the camera sensors (color/IR). The input frames are processed according to the time of day. For instance, IR input is image-enhanced to differentiate between camouflaged intruders from the background. Then input goes through Gaussian Mixture Models (GMM) based change detector to segment foreground (intruder). The foreground object is further cleansed using morphological operations for possible isolated noise and holes. The ICD was tested on three datasets and outperformed top-ranked change detection algorithms.

Path persistence in the cloud

A commonly held belief is that traffic engineering and routing changes are infrequent. However, based on our measurements over a number of years of traffic between data centers in one of the largest cloud provider's networks, we found that it is common for flows to change paths at ten-second intervals or even faster. These frequent path and, consequently, latency variations can negatively impact the performance of cloud applications, specifically, latency-sensitive and geo-distributed applications. Our recent measurements and analysis focused on observing path changes and latency variations between different Amazon aws regions. To this end, we devised a path change detector that we validated using both ad hoc experiments and feedback from cloud networking experts. The results provide three main insights: (1) Traffic Engineering (TE) frequently moves (TCP and UDP) flows among network paths of different latency, (2) Flows experience unfair performance, where a subset of flows between two machines can suffer large latency penalties (up to 32% at the 95th percentile) or excessive number of latency changes, and (3) Tenants may have incentives to selfishly move traffic to low latency classes (to boost the performance of their applications). We showcase this third insight with an example using rsync synchronization. To the best of our knowledge, this is the first paper to reveal the high frequency of TE activity within a large cloud provider's network. Based on these observations, we expect our paper to spur discussions and future research on how cloud providers and their tenants can ultimately reconcile their independent and possibly conflicting objectives. Our data is publicly available for reproducibility and further analysis at http://goo.gl/25BKte.

Discrimination of Internal Faults and Other Transients in an Interconnected System With Power Transformers and Phase Angle Regulators

This study solves the problem of accurate detection of internal faults and classification of transients in a 5-bus interconnected system for Phase Angle Regulators (PAR) and Power Transformers. The analysis prevents mal-operation of differential relays in case of transients other than faults which include magnetizing inrush, sympathetic inrush, external faults with CT saturation, capacitor switching, non-linear load switching, and ferroresonance. A gradient boosting classifier (GBC) is used to distinguish the internal faults from the transient disturbances based on 1.5 cycles of 3-phase differential currents registered by a change detector. After the detection of an internal fault, GBCs are used to locate the faulty unit (Power Transformer, PAR series, or exciting unit) and identify the type of fault. In case a transient disturbance is detected, another GBC classifies them into the six transient disturbances. Five most relevant frequency and time domain features obtained using Information Gain are used to train and test the classifiers. The proposed algorithm distinguishes the internal faults from the other transients with a balanced accuracy of 99.95%. The faulty transformer unit is located with a balanced accuracy of 99.5% and the different transient disturbances are identified with a balanced accuracy of 99.3%. Moreover, the reliability of the scheme is verified for different rating and connection of the transformers involved, CT saturation, and noise levels in the signals. These GBC classifiers can work together with a conventional differential relay and offer a supervisory control over its operation. PSCAD/EMTDC software is used for simulation of the transients and to develop the two and three-winding transformer models for creating the internal faults including inter-turn and inter-winding faults.

Optimizing Data Transmission from IoT Devices Through Weighted Online Data Changing Detectors

The real-time data analysis requires an integrated approach to know the last known state of variables of a concept under monitoring. Thereby, the Internet-of-Thing (IoT) devices have provided alternatives to address distributed data collection strategies. However, the autonomy of IoT devices represents one of the main challenges to implement the collecting strategy. Battery autonomy is affected directly by the energy consumption derived from data transmissions. The Data Stream Processing Strategy (DSPS) is an architecture oriented to the implementation of measurement projects based on a measurement and evaluation framework. Its online processing is guided by the measurement metadata informed from IoT devices associated with a component named Measurement Adapter (MA). This paper presents a new data buffer organization based on measurement metadata articulated with online data filtering to optimize the data transmissions from MA. As contributions, a weighted data change detection approach is incorporated, while a new local buffer based on logical windows is proposed for MA. Also, an articulation among the data buffer, a temporal barrier, and data change detectors is introduced. The proposal was implemented and released on the pabmmCommons library. A discrete simulation on the library is here described to provide initial applicability patterns. The data buffer consumed 568 Kb for monitoring 100 simultaneous metrics. The online estimation of the mean and variance based on the Statistical Process Control consumed 238 ns. However, as a limitation, other scenarios need to be addressed before generalizing results. As future work, new alternatives to filter noise online will be addressed.

Innate Lymphoid Cells in the Airways: Their Functions and Regulators.

Since the airways are constantly exposed to various pathogens and foreign antigens, various kinds of cells in the airways—including structural cells and immune cells—interact to form a precise defense system against pathogens and antigens that involve both innate immunity and acquired immunity. Accumulating evidence suggests that innate lymphoid cells (ILCs) play critical roles in the maintenance of tissue homeostasis, defense against pathogens and the pathogenesis of inflammatory diseases, especially at body surface mucosal sites such as the airways. ILCs are activated mainly by cytokines, lipid mediators and neuropeptides that are produced by surrounding cells, and they produce large amounts of cytokines that result in inflammation. In addition, ILCs can change their phenotype in response to stimuli from surrounding cells, which enables them to respond promptly to microenvironmental changes. ILCs exhibit substantial heterogeneity, with different phenotypes and functions depending on the organ and type of inflammation, presumably because of differences in microenvironments. Thus, ILCs may be a sensitive detector of microenvironmental changes, and analysis of their phenotype and function at local sites may enable us to better understand the microenvironment in airway diseases. In this review, we aimed to identify molecules that either positively or negatively influence the function and/or plasticity of ILCs and the sources of the molecules in the airways in order to examine the pathophysiology of airway inflammatory diseases and facilitate the issues to be solved.

Design of horizontal shear strain instrument for co-estimation of soil stress-strain

This study aimed to design and fabricate the horizontal shear strain transducer (HST) for measuring the soil strain under the agricultural machinery traffic by using a rotary potentiometer which was an angular strain change detector. The angular movable mechanism was calibrated and evaluated the characteristics of the instrument in the laboratory, and found that the non-linearity and the hysteresis were 4.56 and 4.42%, respectively. Furthermore, the strain measurement device was buried in the field to compare the shear strain at varying tractor velocities (5, 6.5 and 8 km/h). For all vehicle speed at the first passage, the shear strain and soil bulk density obviously increased, whereas the final moisture content decreased; as a result, the lower velocity caused higher shear strain.

Robust Low-Rank Change Detection for Multivariate SAR Image Time Series

This article derives a new change detector for multivariate synthetic aperture radar (SAR) image time series (ITS). Classical statistical change detection methodologies based on covariance matrix analysis are usually built upon the Gaussian assumption, as well as an unstructured signal model. Both of these hypotheses may be inaccurate for high-dimension/resolution images, where the noise can be heterogeneous (non-Gaussian) and where the relevant signals usually lie in a low-dimensional subspace (low-rank structure). These two issues are tackled by proposing a new generalized likelihood ratio test based on a robust (compound Gaussian) low-rank (structured covariance matrix) model. The interest of the proposed detector is assessed on two SAR-ITS set from UAVSAR.

Change Detection Based on Low-Level to High-Level Features Integration With Limited Samples

Detailed land cover change in multitemporal images is an important application for earth science. Many techniques have been proposed to solve this problem in different ways. However, accurately identifying changes still remains a challenge due to the difficulties in describing the characteristics of various change categories by using single-level features. In this article, a multilevel feature representation framework was designed to build robust feature set for complex change detection task. First, four different levels of information from low level to high level, including pixel-level, neighborhood-level, object-level, and scene-level features, were extracted. Through the operation of extracting different level features from multitemporal images, the differences between them can be described comprehensively. Second, multilevel features were fused to reduce the dimension and then used as the input for supervised change detector with initial limited labels. Finally, for reducing the labeling cost and improving the change detection results simultaneously, active learning was conducted to select the most informative samples for labeling, and this step together with the previous steps were iteratively conducted to improve the results in each round. Experimental results of three pairs of real remote sensing datasets demonstrated that the proposed framework outperformed the other state-of-the-art methods in terms of accuracy. Moreover, the influences of scene scale for high-level semantic features in the proposed approach on change detection performance were also analyzed and discussed.

Condition Monitoring for Single-Rotor Wind Turbine Load Sensors in the Full-Load Region

Incipient faults on blade load sensors can impede the sound performance of a wind turbine, leading to increasing loads over time and severe blade degradation. As such, knowledge of the blade load sensors’ health is essential for ensuring effective load reduction by means of individual pitch control. This paper presents a condition monitoring strategy for the blade load sensors based on estimation of the loads acting on the rotor blades in the full-load region. Fault detection is achieved via appropriate residual generators, the statistical properties of which are used to design change detectors robust against measurement noise and wind field stochasticity. Specifically, a Generalized Likelihood Ratio Test for the t-LocationScale distribution is developed for ensuring robust detection of sensor blade faults. The proposed method is evaluated in a high-fidelity simulator under non-uniform wind scenarios. The simulation results show that detection of multiplicative faults on the blade load sensors is achieved even in absence of knowledge of the local wind speed.

Detection of Wind Turbines in Intertidal Areas Using SAR Polarimetry

The detection of wind turbines in a strong clutter background is analyzed at variance of polarimetric synthetic-aperture radar (SAR) configurations. The area of interest is the intertidal zone near Jiangsu, China and two detectors are used, the polarimetric notch filter (PNF) and a change detector that optimizes the ratio between covariance matrices. The detection performance is quantitatively analyzed using the receiver operating characteristic (ROC) curve, while the scattering mechanisms that characterize wind turbines are analyzed using the Yamaguchi decomposition. Experimental analysis shows that: 1) wind turbines result in a nontrivial scattering mechanism and 2) full-polarimetric measurements achieve the best detection performance independently of the two detectors.

Kernel Anomalous Change Detection for Remote Sensing Imagery

Anomalous change detection (ACD) is an important problem in remote sensing image processing. Detecting not only pervasive but also anomalous or extreme changes has many applications for which methodologies are available. This paper introduces a nonlinear extension of a full family of anomalous change detectors. In particular, we focus on algorithms that utilize Gaussian and elliptically contoured (EC) distribution and extend them to their nonlinear counterparts based on the theory of reproducing kernels' Hilbert space. We illustrate the performance of the kernel methods introduced in both pervasive and ACD problems with real and simulated changes in multispectral and hyperspectral imagery with different resolutions (AVIRIS, Sentinel-2, WorldView-2, and Quickbird). A wide range of situations is studied in real examples, including droughts, wildfires, and urbanization. Excellent performance in terms of detection accuracy compared to linear formulations is achieved, resulting in improved detection accuracy and reduced false-alarm rates. Results also reveal that the EC assumption may be still valid in Hilbert spaces. We provide an implementation of the algorithms as well as a database of natural anomalous changes in real scenarios http://isp.uv.es/kacd.html.

A Reliable Mixed-Norm-Based Multiresolution Change Detector in Heterogeneous Remote Sensing Images

Analysis of heterogeneous remote sensing image is a challenging and complex problem due to the fact that the local statistics of the data to be processed can be radically different. In this article, we present a novel and reliable unsupervised change detection (CD) method to analyze heterogeneous remotely sensed image pairs. The proposed method is based on an imaging modality-invariant operator that detects at different scale levels the differences in terms of high-frequency pattern of each structural region existing in the two heterogeneous satellite images. First, this new detector is based upon a dual-norm formulation that makes our underlying CD estimation particularly robust in terms of a sensitivity/specificity tradeoff. Second, the detection process, embedded in a multiresolution framework, allows us to estimate a robust similarity or difference map that is then filtered out by a superpixel-based spatially adaptive filter to further increase its reliability against noise. Finally, changes are then identified from this similarity map by a simple binary clustering process that also takes into account the spatial contextual information around each pixel. Experimental results involving different types of heterogeneous remotely sensed image pairs confirm the robustness of the proposed approach.

A generic deviance detection principle for cortical On/Off responses, omission response, and mismatch negativity

Neural responses to sudden changes can be observed in many parts of the sensory pathways at different organizational levels. For example, deviants that violate regularity at various levels of abstraction can be observed as simple On/Off responses of individual neurons or as cumulative responses of neural populations. The cortical deviance-related responses supporting different functionalities (e.g., gap detection, chunking, etc.) seem unlikely to arise from different function-specific neural circuits, given the relatively uniform and self-similar wiring patterns across cortical areas and spatial scales. Additionally, reciprocal wiring patterns (with heterogeneous combinations of excitatory and inhibitory connections) in the cortex naturally speak in favor of a generic deviance detection principle. Based on this concept, we propose a network model consisting of reciprocally coupled neural masses as a blueprint of a universal change detector. Simulation examples reproduce properties of cortical deviance-related responses including the On/Off responses, the omitted-stimulus response (OSR), and the mismatch negativity (MMN). We propose that the emergence of change detectors relies on the involvement of disinhibition. An analysis of network connection settings further suggests a supportive effect of synaptic adaptation and a destructive effect of N-methyl-d-aspartate receptor (NMDA-r) antagonists on change detection. We conclude that the nature of cortical reciprocal wiring gives rise to a whole range of local change detectors supporting the notion of a generic deviance detection principle. Several testable predictions are provided based on the network model. Notably, we predict that the NMDA-r antagonists would generally dampen the cortical Off response, the cortical OSR, and the MMN.

Change detection in high-resolution images based on feature importance and ensemble method

A robust framework for change detection in high-resolution (HR) images which takes into account feature importance of morphological attribute profiles (APs) is proposed. Although APs allow the discriminative extraction of geometrical features, the selection of optimal filtering thresholds is always a challenging task. In order to address this problem, the importances of multi-attribute APs, which are generated by ranges of thresholds, are calculated in advance. The APs with higher importances are chosen as input features for change detectors. Furthermore, owing to the possible incomplete feature importance model caused by randomness and small size of selected training samples through random forest, an ensemble method is conducted by integrating different results to enhance the stability of the final output. Experimental results obtained by two pairs of bi-temporal HR images demonstrate the effectiveness of the proposed approach.

Detection of inconspicuous power quality disturbances through step changes in rms voltage profile

Power quality disturbances commonly observed in power systems have been studied for decades, resulting in numerous algorithms for detecting the events that affect the voltage and/or current waveforms. However, a considerable amount of disturbances is not visually observable in the raw waveforms, especially switching operations. These events must be detected through an alternative feature, such as abrupt variations in the root-mean-square (rms) voltage profile. This study examines the methods commonly used for detecting power quality disturbances in the waveform or rms voltage profile domains and identifies their limitations. Afterwards, a novel step change detector is proposed based on a modified median filter and rms voltage gradient values to overcome the deficiencies of the existing methods. The effectiveness of the proposed method is assessed by applying it to both simulated and field data. This assessment shows that the method detects all switching events with no false positives for the datasets under analysis.

Multiple changepoint detection in categorical data streams

The need for efficient tools is pressing in the era of big data, particularly in streaming data applications. As data streams are ubiquitous, the ability to accurately detect multiple changepoints, without affecting the continuous flow of data, is an important issue. Change detection for categorical data streams is understudied, and existing work commonly introduces fixed control parameters while providing little insight into how they may be chosen. This is ill-suited to the streaming paradigm, motivating the need for an approach that introduces few parameters which may be set without requiring any prior knowledge of the stream. This paper introduces such a method, which can accurately detect changepoints in categorical data streams with fixed storage and computational requirements. The detector relies on the ability to adaptively monitor the category probabilities of a multinomial distribution, where temporal adaptivity is introduced using forgetting factors. A novel adaptive threshold is also developed which can be computed given a desired false positive rate. This method is then compared to sequential and nonsequential change detectors in a large simulation study which verifies the usefulness of our approach. A real data set consisting of nearly 40 million events from a computer network is also investigated.

Towards precise one-way fiber-based frequency dissemination using phase-sensitive amplification.

In this Letter, we conceptually demonstrate the potential of a phase-sensitive amplifier to operate as an active detector of stochastic phase changes in fiber-based frequency dissemination systems with two orders of magnitude better sensitivity than state-of-the-art one-way systems relying on two-wavelength dissemination schemes. Theoretical and experimental analyses show that these stochastic phase changes (caused by environmental changes, e.g., due to temperature) can be detected with high sensitivity via optical phase comparison performed within the phase-sensitive amplifier. Experimental results are in close agreement with theoretical predictions showing that phase-sensitive amplifiers may find a niche application in metrology, with potential to significantly improve one-way fiber-based frequency dissemination systems.

Improved Dynamic PSO-Based Algorithm for Critical Spare Parts Supply Optimization Under (T, S) Inventory Policy

This paper investigates the multi-period and multi-echelon spare parts supply optimization problem. The periodic review order-up-to-S inventory policy is considered to determine the supply timing and the spare parts consumption. The spare parts consumption in lead time of each customer is calculated by the process of spare parts supply. A dynamic optimization mathematical model is developed to find the optimal supply scheme in each period with the objective of minimum total costs. In the dynamic optimization model, the parameters are changed with time. An improved particle swarm optimization algorithm is proposed to solve this dynamic problem. In the proposed algorithm, a change detector and a change response strategy are used to handle the change of optimization environment. We also adopt a self-adaption neighborhood search strategy to improve the ability of exploration and exploitation of the algorithm. Numerical experiment shows that the proposed strategy can improve the performance of algorithm greatly, and the model and solution method in this paper can be applied to optimize the spare parts supply in various scenarios.

A Novel Fault-Tolerant Navigation and Positioning Method with Stereo-Camera/Micro Electro Mechanical Systems Inertial Measurement Unit (MEMS-IMU) in Hostile Environment.

Visual odometry (VO) is a new navigation and positioning method that estimates the ego-motion of vehicles from images. However, VO with unsatisfactory performance can fail severely in hostile environment because of the less feature, fast angular motions, or illumination change. Thus, enhancing the robustness of VO in hostile environment has become a popular research topic. In this paper, a novel fault-tolerant visual-inertial odometry (VIO) navigation and positioning method framework is presented. The micro electro mechanical systems inertial measurement unit (MEMS-IMU) is used to aid the stereo-camera, for a robust pose estimation in hostile environment. In the algorithm, the MEMS-IMU pre-integration is deployed to improve the motion estimation accuracy and robustness in the cases of similar or few feature points. Besides, a dramatic change detector and an adaptive observation noise factor are introduced, tolerating and decreasing the estimation error that is caused by large angular motion or wrong matching. Experiments in hostile environment showing that the presented method can achieve better position estimation when compared with the traditional VO and VIO method.