Search results

The purpose of this paper is to address the problem of profiling human behaviour patterns captured in surveillance videos for the application of online normal behaviour recognition and anomaly detection.

A novel framework is developed for automatic behaviour profiling and online anomaly detection without any manual labeling of the training dataset.

Experimental results demonstrate the effectiveness and robustness of the authors' approach using noisy and sparse datasets collected from one real surveillance scenario.

To discover the topics, co‐clustering topic model not only captures the correlation between words, but also models the correlations between topics. The major difference between the conventional co‐clustering algorithms and the proposed CCMT is that CCMT shows a major improvement in terms of recall, i.e. interpretability.

Visual inspection and human judgement form the cornerstone of daily operations and maintenance (O&M) services activities carried out by facility managers nowadays. Recent advances in technologies such as building information modelling (BIM), distributed sensor networks, augmented reality (AR) technologies and digital twins present an immense opportunity to radically improve the way daily O&M is conducted. This paper aims to describe the development of an AR-supported automated environmental anomaly detection and fault isolation method to assist facility managers in addressing problems that affect building occupants’ thermal comfort.

The developed system focusses on the detection of environmental anomalies related to the thermal comfort of occupants within a building. The performance of three anomaly detection algorithms in terms of their ability to detect indoor temperature anomalies is compared. Based on the fault tree analysis (FTA), a decision-making tree is developed to assist facility management (FM) professionals in identifying corresponding failed assets according to the detected anomalous symptoms. The AR system facilitates easy maintenance by highlighting the failed assets hidden behind walls/ceilings on site to the maintenance personnel. The system can thus provide enhanced support to facility managers in their daily O&M activities such as inspection, recording, communication and verification.

Taking the indoor temperature inspection as an example, the case study demonstrates that the O&M management process can be improved using the proposed AR-enhanced inspection system. Comparative analysis of different anomaly detection algorithms reveals that the binary segmentation-based change point detection is effective and efficient in identifying temperature anomalies. The decision-making tree supported by FTA helps formalise the linkage between temperature issues and the corresponding failed assets. Finally, the AR-based model enhanced the maintenance process by visualising and highlighting the hidden failed assets to the maintenance personnel on site.

The originality lies in bringing together the advances in augmented reality, digital twins and data-driven decision-making to support the daily O&M management activities. In particular, the paper presents a novel binary segmentation-based change point detection for identifying temperature anomalous symptoms, a decision-making tree for matching the symptoms to the failed assets, and an AR system for visualising those assets with related information.

Course evaluations are formative and are used to evaluate learnings of the students for a course. Anomalies in the evaluation process can lead to a faulty educational outcome. Learning analytics and educational data mining provide a set of techniques that can be conveniently applied to extensive data collected as part of the evaluation process to ensure remedial actions. This study aims to conduct an experimental research to detect anomalies in the evaluation methods.

Experimental research is conducted with scientific approach and design. The researchers categorized anomaly into three categories, namely, an anomaly in criteria assessment, subject anomaly and anomaly in subject marks allocation. The different anomaly detection algorithms are used to educate data through the software R, and the results are summarized in the tables.

The data points occurring in all algorithms are finally detected as an anomaly. The anomaly identifies the data points that deviate from the data set’s normal behavior. The subject which is consistently identified as anomalous by the different techniques is marked as an anomaly in evaluation. After identification, one can drill down to more details into the title of anomalies in the evaluation criteria.

This paper proposes an analytical model for the course evaluation process and demonstrates the use of actionable analytics to detect anomalies in the evaluation process.

The defect detection problem of color-patterned fabric is still a huge challenge due to the lack of manual defect labeling samples. Recently, many fabric defect detection algorithms based on feature engineering and deep learning have been proposed, but these methods have overdetection or miss-detection problems because they cannot adapt to the complex patterns of color-patterned fabrics. The purpose of this paper is to propose a defect detection framework based on unsupervised adversarial learning for image reconstruction to solve the above problems.

The proposed framework consists of three parts: a generator, a discriminator and an image postprocessing module. The generator is able to extract the features of the image and then reconstruct the image. The discriminator can supervise the generator to repair defects in the samples to improve the quality of image reconstruction. The multidifference image postprocessing module is used to obtain the final detection results of color-patterned fabric defects.

The proposed framework is compared with state-of-the-art methods on the public dataset YDFID-1(Yarn-Dyed Fabric Image Dataset-version1). The proposed framework is also validated on several classes in the MvTec AD dataset. The experimental results of various patterns/classes on YDFID-1 and MvTecAD demonstrate the effectiveness and superiority of this method in fabric defect detection.

It provides an automatic defect detection solution that is convenient for engineering applications for the inspection process of the color-patterned fabric manufacturing industry. A public dataset is provided for academia.

This study aims to address the challenge of training a detection model for the robot to detect the abnormal samples in the industrial environment, while abnormal patterns are very rare under this condition.

The authors propose a new model with double encoder–decoder (DED) generative adversarial networks to detect anomalies when the model is trained without any abnormal patterns. The DED approach is used to map high-dimensional input images to a low-dimensional space, through which the latent variables are obtained. Minimizing the change in the latent variables during the training process helps the model learn the data distribution. Anomaly detection is achieved by calculating the distance between two low-dimensional vectors obtained from two encoders.

The proposed method has better accuracy and F1 score when compared with traditional anomaly detection models.

A new architecture with a DED pipeline is designed to capture the distribution of images in the training process so that anomalous samples are accurately identified. A new weight function is introduced to control the proportion of losses in the encoding reconstruction and adversarial phases to achieve better results. An anomaly detection model is proposed to achieve superior performance against prior state-of-the-art approaches.

The purpose of this paper is to ensure the stable operation of satellites in orbit and to assist ground personnel in continuously monitoring the satellite telemetry data and finding anomalies in advance, which can improve the reliability of satellite operation and prevent catastrophic losses.

This paper proposes a deep auto-encoder (DAE) satellite anomaly advance warning framework for satellite telemetry data. Firstly, this study performs grey correlation analysis, extracts important feature attributes to construct feature vectors and builds the variational auto-encoder with bidirectional long short-term memory generative adversarial network discriminator (VAE/BLGAN). Then, the Mahalanobis distance is used to measure the reconstruction score of input and output. According to the periodic characteristic of satellite operation, a dynamic threshold method based on periodic time window is proposed. Satellite health monitoring and advance warning are achieved using reconstruction scores and dynamic thresholds.

Experiment results indicate DAE methods can probe that satellite telemetry data appear abnormal, trigger a warning before the anomaly occurring and thus allow enough time for troubleshooting. This paper further verifies that the proposed VAE/BLGAN model has stronger data learning ability than other two auto-encoder models and is sensitive to satellite monitoring data.

This paper provides a DAE framework to apply in the field of satellite health monitoring and anomaly advance warning. To the best of the authors’ knowledge, this is the first paper to combine DAE methods with satellite anomaly detection, which can promote the application of artificial intelligence in spacecraft health monitoring.

The purpose of this paper is to propose a structured multilevel system that will distinguish the anomalies present in different online social networks (OSN).

Author first reviewed the related work, and then, the research model designed was explained. Furthermore, the details regarding Levels 1 and 2 were narrated.

By using the proposed technique, F_Score obtained for Twitter and Facebook data set was 96.22 and 94.63, respectively.

Four data sets were used for the experiment and the acquired outcomes demonstrate enhancement over the current existing frameworks.

This paper designed a multilevel framework that can be used to detect the anomalies present in the OSN.

The purpose of this paper is to detect the occurrence of anomaly and fault in a spacecraft, investigate various tendencies of telemetry parameters and evaluate the operation state of the spacecraft to monitor the health of the spacecraft.

This paper proposes a data-driven method (empirical mode decomposition-sample entropy-principal component analysis [EMD-SE-PCA]) for monitoring the health of the spacecraft, where EMD is used to decompose telemetry data and obtain the trend items, SE is utilised to calculate the sample entropies of trend items and extract the characteristic data and squared prediction error and statistic contribution rate are analysed using PCA to monitor the health of the spacecraft.

Experimental results indicate that the EMD-SE-PCA method could detect characteristic parameters that appear abnormally before the anomaly or fault occurring, could provide an abnormal early warning time before anomaly or fault appearing and summarise the contribution of each parameter more accurately than other fault detection methods.

The proposed EMD-SE-PCA method has high level of accuracy and efficiency. It can be used in monitoring the health of a spacecraft, detecting the anomaly and fault, avoiding them timely and efficiently. Also, the EMD-SE-PCA method could be further applied for monitoring the health of other equipment (e.g. attitude control and orbit control system) in spacecraft and satellites.

The paper provides a data-driven method EMD-SE-PCA to be applied in the field of practical health monitoring, which could discover the occurrence of anomaly or fault timely and efficiently and is very useful for spacecraft health diagnosis.

Transaction monitoring system set up by financial institutions is one of the most used ways to track money laundering and terrorist financing activities. While being effective to a large extent, the system generates very high false positives. With evolving patterns of financial transactions, it also needs effective mechanism for scenario fine-tuning. The purpose of this paper is to highlight quantitative method for optimizing scenarios in money laundering context. While anomaly detection and unsupervised learning can identify huge patterns of false negatives, that can reveal new patterns, for existing scenarios, business generally rely on judgment/data analysis-based threshold finetuning of existing scenario. The objective of such exercises is productivity rate enhancement.

In this paper, the authors propose an approach called linear/non-linear optimization on threshold finetuning. This traditional operations research technique has been often used for many optimization problems. Current problem of threshold finetuning for scenario has two key features that warrant linear optimization. First, scenario-based suspicious transaction reporting (STR) cases and overall customer level catch rate has a very high overlap, i.e. more than one scenario captures same customer with different degree of abnormal behavior. This implies that scenarios can be better coordinated to catch more non-overlapping customers. Second, different customer segments have differing degree of transaction behavior; hence, segmenting and then reducing slack (redundant catch of suspect) can result in better productivity rate (defined as productive alerts divided by total alerts) in a money laundering context.

Theresults show that by implementing the optimization technique, the productivity rate can be improved. This is done through two drivers. First, the team gets to know the best possible combination of threshold across scenarios for maximizing the STR observations better coverage of STR – fine-tuned thresholds are able to better cover the suspected transactions as compared to traditional approaches. Second, there is reduction of redundancy/slack margins on thresholds, thereby improving the overall productivity rate. The experiments focused on six scenario combinations, resulted in reduction of 5.4% of alerts and 1.6% of unique customers for same number of STR capture.

The authors propose an approach called linear/non-linear optimization on threshold finetuning, as very little work is done on optimizing scenarios itself, which is the most widely used practice to monitor enterprise-wide anti-money laundering solutions. This proves that by adding a layer of mathematical optimization, financial institutions can additionally save few million dollars, without compromising on their STR capture capability. This hopefully will go a long way in leveraging artificial intelligence for further making financial institutions more efficient in controlling financial crimes and save some hard-earned dollars.

In novelty detection, the autoencoder based image reconstruction strategy is one of the mainstream solutions. The basic idea is that once the autoencoder is trained on normal data, it has a low reconstruction error on normal data. However, when faced with complex natural images, the conventional pixel-level reconstruction becomes poor and does not show the promising results. This paper aims to provide a new method for improving the performance of novelty detection based autoencoder.

To solve the problem that conventional pixel-level reconstruction cannot effectively extract the global semantic information of the image, a novel model with the combination of attention mechanism and self-supervised learning method is proposed. First, an auxiliary task, reconstruct rotated image, is set to enable the network to learn global semantic feature information. Then, the channel attention mechanism is introduced to perform adaptive feature refinement on the intermediate feature map to optimize the correspondingly passed feature map.

Experimental results on three public data sets show that the proposed method has potential performance for novelty detection.

This study explores the ability of self-supervised learning methods and attention mechanism to extract features on a single class of images. In this way, the performance of novelty detection can be improved.