Abnormal Detection Method Research Articles

Ensemble learning-based abnormality diagnosis in wrist skeleton radiographs using densenet variants voting

Almost one out of five people, including children, suffers from musculoskeletal disorders. It is the second leading cause of disability worldwide. It affects the musculoskeletal system’s major areas, represented by the shoulder, forearm, and wrist. It causes severe pain, joint noises, and disability. To detect the abnormality, the radiologist analyzes the patient’s anatomy through X-rays of different views and projections. To automatically diagnose the abnormality in the musculoskeletal system is a challenging task. Previously, various researchers detected the abnormality in the musculoskeletal system from radiographic images by using several deep learning techniques. They used a capsule network, 169-layer convolutional neural network, and group normalized convolutional neural network in musculoskeletal abnormality detection. However, to propose methods for improving abnormality detection, further work needs to be done because the accuracy of the conventional methods is far away from 90%. This paper presents an ensemble learning-based classification system for detecting abnormality in wrist radiographs. Tags in radiographs may result in learning noisy features hence reducing the performance. Therefore, tags are segmented and removed using UNet trained on the annotated ground truths. Segmented images are then used for voting-based diagnosis. The simulation results show that the proposed methodology improves testing accuracy by 1.5%-4.5% compared to the available wrist abnormality detection methods. The proposed methodology can be used for any kind of musculoskeletal abnormality detection.

Abnormal Target Detection Method in Hyperspectral Remote Sensing Image Based on Convolution Neural Network.

Abnormal target detection in hyperspectral remote sensing image is one of the hotspots in image research. The image noise generated in the detection process will lead to the decline of the quality of hyperspectral remote sensing image. In view of this, this paper proposes an abnormal target detection method of hyperspectral remote sensing image based on the convolution neural network. Firstly, the deep residual learning network model has been used to remove the noise in hyperspectral remote sensing image. Secondly, the spatial and spectral features of hyperspectral remote sensing images were used to optimize the clustering dictionary, and then the image segmentation containing target information is completed. Finally, the image was input into the deep convolution neural network with a dual classifier, and the network detects the abnormal target in the image. The test results of this algorithm show that the structural similarity of the denoised image is higher than 0.86, which shows that this method has good noise reduction performance, image details will not damage, segmentation effect is good, and it can obtain high-definition target image information and accurately detect abnormal targets in the image.

Probing individual-level structural atrophy in frontal glioma patients.

Although every glioma patient varies in tumor size, location, histological grade and molecular biomarkers, non-tumoral morphological abnormalities are commonly detected by a statistical comparison among patient groups, missing the information of individual morphological alterations. In this study, we introduced an individual-level structural abnormality detection method for glioma patients and proposed several abnormality indexes to depict individual atrophy patterns. Forty-five patients with a glioma in the frontal lobe and fifty-one age-matched healthy controls participated in the study. Individual structural abnormality maps (SAM) were generated using patients' preoperative T1 images, by calculating the degree of deviation of voxel volume in each patient with the normative model built from healthy controls. Based on SAM, a series of individual abnormality indexes were computed, and their relationship with glioma characteristics was explored. The results demonstrated that glioma patients showed unique non-tumoral atrophy patterns with overlapping atrophy regions mainly located at hippocampus, parahippocampus, amygdala, insula, middle temporal gyrus and inferior temporal gyrus, which are closely related to the human cognitive functions. The abnormality indexes were associated with several molecular biomarkers including isocitrate dehydrogenase (IDH) mutation, 1p/19q co-deletion and telomerase reverse transcriptase (TERT) promoter mutation. Our study provides an effective way to access the individual-level non-tumoral structural abnormalities in glioma patients, which has the potential to significantly improve individualized precision medicine.

Detection of Multicamera Pedestrian Trajectory Outliers in Geographic Scene

The abnormal detection of moving objects in intelligent video surveillance system plays an important role in early warning for man-made disasters. However, the current abnormal detection methods cannot effectively perceive the cross-camera abnormal movements of video objects. The main reason is that the existing methods ignore the spatial relationship between the fields of view of different cameras and blind areas among the fields of view. This condition prevents them to effectively infer and analyze the cross-camera movements of video objects combined with geospatial information. This paper proposes the detection of multicamera pedestrian trajectory outliers in geographic scene to address this problem. This approach first spatializes the video object trajectory and then realizes trajectory vectorization by extracting trajectory points with equal time difference. The position trajectory outliers are detected by constructing isolation forest and scoring trajectory vectors, and the velocity trajectory outliers are identified through vectors’ neighborhood comparison. Related experiments show that our method can effectively improve the efficiency and accuracy of detecting trajectory outliers, which can enhance the early warning capability of video surveillance systems for man-made disasters.

The abnormal traffic detection scheme based on PCA and SSH

Network abnormal traffic detection can monitor the network environment in real time by extracting and analysing network traffic characteristics, and plays an important role in network security protection. In order to solve the problems that the existing detection methods cannot fully learn the spatio-temporal characteristics of data, the classification accuracy is not high, and the detection time and accuracy are susceptible to the influence of redundant data in the sample. Thus, this paper proposes a network abnormal detection method (PCSS) integrating principal component analysis (PCA) and single-stage headless face detector algorithms (SSH). PCSS applies the PCA algorithm to the data preprocessing to eliminate the interference of redundant data. At the same time, PCSS also combines feature fusion and SSH to enhance the feature extraction of unclear features data, and effectively improve the detection speed and accuracy. Simulation experiments based on IDS2017 and IDS2012 data sets are carried out in this paper. Experimental results show that PCSS is obviously superior to other detection models in detection speed and accuracy, which provides a new method for efficiently detecting traffic attacks.

An abnormal user login behavior detection method of industrial control system based on multi-dimensional probability analysis

Aiming at the poor detection performance of user abnormal login behavior in the complex environment of industrial control system, a new method based on multi-dimensional probability analysis is proposed. First of all, login time, source, destination and other login parameters are collected. Secondly, the abnormal probability of login parameters is calculated. Thirdly, the login transfer probability among multiple destination hosts is calculated. Finally, through the comprehensive analysis of the above three probabilities, an abnormal login behavior detection model based on multi-dimensional probability analysis is constructed. Experiments show that this method can effectively identify the abnormal login behavior of users caused by various types of network attacks.

Abnormal Data Detection and Identification Method of Distribution Internet of Things Monitoring Terminal Based on Spatiotemporal Correlation

As an important part of the ubiquitous power Internet of Things, the distribution Internet of Things can further improve the automation and informatization level of the distribution network. The reliability of the measurement data of the low-voltage terminal unit, as the sensing unit of the sensing layer of the distribution Internet of Things, has a great impact on the fault processing and advanced applications of the distribution Internet of Things. The self-check and the equipment working status monitoring of the main station of the low-voltage terminal unit struggle to identify the abnormality of measurement data. Aiming at this problem, an abnormal data detection and identification recognition method of a distribution Internet of Things monitoring terminal is proposed on the basis of spatiotemporal correlation. First, using the temporal correlation of monitoring terminal data, the proposed composite temporal series similarity measurement criterion is used to calculate the distance matrix between data, and the abnormal data detection is realized via combination with the improved DBSCAN algorithm. Then, using the spatial correlation of the data of the terminal unit, the geometric features of the spatial cross-correlation coefficient of the terminal nodes are extracted as the input of the cascaded fuzzy logic system to identify the abnormal source. Lastly, the effectiveness of the method is verified by a practical example.

Laplacian matrix graph for anomaly target detection in hyperspectral images

To improve the accuracy of abnormal target detection in hyperspectral images, an abnormal target detection method based on Laplacian matrix graph (LGD) is proposed. The method makes full use of the spatial and spectral information of hyperspectral abnormal targets by constructing the full-connection graph and the nearest neighbour matrix obtained by the Gaussian kernel function. In the graph, the total variation of the graph signal calculated by the Laplacian matrix is taken as the evaluation function to judge the abnormal target, so as to realize the detection of abnormal pixels. It avoids the matrix inversion calculation that must be carried out in the conventional detection algorithms and the complexity is reduced. Compared with other detection algorithms on three different data sets, the experiment results show that the proposed algorithm presents obvious advantages in detection accuracy and excellent utility in practice.

Ship Abnormal Behavior Detection Method Based on Optimized GRU Network

Ship abnormal behavior detection is an essential part of maritime supervision. It can assist maritime departments to conduct real-time supervision on a certain sea area, avoid ship risks, and improve the efficiency of sea area supervision. Given the problems of complex detection methods, poor detection effectiveness, and low detection accuracy, a Gated Recurrent Unit (GRU) was proposed for ship abnormal behavior detection. Under the premise of introducing the attention mechanism into a GRU, the optimal GRU structure parameters were obtained through the intelligent algorithm to perform deeper feature extraction and train the ship abnormal behavior based on the optimized GRU neural network, so as to realize the detection and recognition of the trajectory data to be measured. Finally, based on the public data set and the trajectory data of the inward and outward ports of ships issued by Nanjing Section, Jiangsu Maritime Bureau, the TensorFlow frame was used to establish an abnormal behavior detection model. The simulation results demonstrated that the abnormal behavior detection model shortened the abnormal detection time. The abnormal behavior detection model used in the detection of ship abnormal behavior enhanced the accuracy and stability of the abnormal behavior identification and verified the validity and superiority of this method.

Abnormal data detection of guidance angle based on SMP-SVDD for seeker

The accuracy of the pitch angle deviation directly affects the guidance accuracy of the laser seeker. During the guidance process, the abnormal pitch angle deviation data will be produced when the seeker is affected by interference sources. In this paper, a new abnormal data detection method based on Smooth Multi-Kernel Polarization Support Vector Data Description (SMP-SVDD) is proposed. In the proposed method, the polarization value is used to determine the weight of the multi-kernel combination coefficient to obtain the multi-kernel polarization function, in which the particle swarm optimization is used to find the optimal kernels for higher detection accuracy. Besides, by using smoothing mechanism, the constrained quadratic programming problem is translated to be smooth and differentiable. Then, this problem can be solved by the conjugate gradient method, which could reduce the computational complexity. In experimental section, abundant simulation experiments were designed and the experimental results verify that the proposed SMP-SVDD method could achieve higher detection accuracy and low computational cost compared with different detection methods in different guidance stages.

Spatial Contrastive Learning for Anomaly Detection and Localization

With the development of deep learning, abnormal detection methods have been widely presented to improve performances in various applications, including visual inspection systems. However, there remains difficult to be directly applied to real-world applications, which often include the lack of abnormal samples and diversity. This paper proposes contra embedding that adopts progressive autoencoder with contrastive learning to address these difficulties. The autoencoder is trained progressively to reproduce the details of the original images, and modified CutPaste augmentation helps to learn to recover normal images. Especially, contrastive learning based on normal embedding vectors effectively reduces false positives caused by the autoencoder. The proposed method is also helpful when normal data have complex shapes, sizes, and colors. In experiments, MVTec AD dataset is used to show the generalization ability of the proposed method in various real-world applications. It achieves over 98.0% AUROCs in detection and 97.7% AUROCs in the localization, respectively, without using the ImageNet pre-trained model as in previous methods.

Study of the abnormal vibration detection method for F type support information board using autoencoder

This study concerns a method for evaluating the structural anomaly of F type support information board from acceleration measurements. Acceleration of the top of supports columns were measure about three years. The evaluation f structural anomaly is performed by the autoencoder method．In the method training of the autoencoder is conducted from the normal conditions, and diagnosis of the structural is conducted by estimation error from the data neural network． The diagnoses are performed to the information board at two sites．Data of site B，which has unstable boundary，has high fluctuation to the natural frequency and constant fluctuation to the average of the natural frequency．Data of site A has stable natural frequency．Additionally site B shows a large change in condition after the typhoon passed, and the autoencoder was able to show a significant change in error value. At site A, the values were stable, and although there were temporary fluctuations, the results were not problematic in the long term. Therefore, the method of structural variation detection using the autoencoder is a practical method.

SSC-KA: A Framework of Space Situational Knowledge Acquisition for Status-Cognition of Satellites

With the rapid development of space technology, the environment of the space domain has become more and more complex and changeable, which brought great difficulties in cognition of space domain activity. As space domain awareness (SDA) required, any relevant information and knowledge from various sources are needed as much as possible, while all of those can be sorted and integrated for effective cognition of space objects, including the cognition of their status. This paper proposes a knowledge integration framework (SSC-KA) designed for the cognition of space target and its status. In the framework (SSC-KA), open-source information and data acquired from multi-kinds of sensors are sent into parallel channels, and then processed by the algorithm this paper designed into sequence data as the channels’ output. Furthermore, the rules and four statuses defined in this paper can be judged for the anomaly detection of satellites. Based on the space domain knowledge acquisition framework of SSC-KA, this paper describes a complete abnormal state detection method for satellites step by step through multi-level feature engineering. Therefore, the method is used to analyze four different statuses of satellites in this paper, to verify the validity and feasibility of the application of the method in the cognition of spatial events, thus laying the foundation for the cognition of Space Domain Awareness.

Introduction of Abnormal Combustion in Hydrogen Internal Combustion Engines and the Detection Method

As a clean, environmentally friendly and renewable energy source, hydrogen as an alternative engine fuel can greatly reduce atmospheric pollution and alleviate the shortage of oil resources, and is the most promising alternative fuel for vehicles among new fuels. However, due to its fast combustion rate and wide ignition limit, hydrogen often shows abnormal combustion phenomena (such as pre-ignition, backfire and knock), when it is used in the engine, thus affecting the performance and normal use of engines. In this paper, the advantages and disadvantages of hydrogen as an alternative fuel for the engine are summarized according to the characteristics of hydrogen. On this basis, the mechanism, influence factors and harm of abnormal combustion in the hydrogen internal combustion engine are analyzed and summarized, which provides a theoretical basis for solving abnormal combustion problems. Finally, several commonly used abnormal combustion detection methods are summarized.

An Abnormal Intrusion Detection Method Based on Self-organizing Model

An Abnormal Intrusion Detection Method Based on Self-organizing Model

Deep learning for in-situ powder stream fault detection in directed energy deposition process

In the directed energy deposition (DED) process, stable powder feeding through nozzles is crucial. However, abnormal powder feeding due to nozzle clogging or adhesion frequently occurs, which can cause severe defects or even lead to a process failure. Nevertheless, the abnormal powder feeding detection is challenging because the powder stream variance is subtle, and the sensing area is under a harsh environment with bright melt pool light emission and spatter ejection. This study proposes the abnormal powder feeding detection method based on a deep learning technique using melt pool images. The trained model by training dataset collected through a coaxial camera with deliberately installed clogged nozzle can discriminate between normal and abnormal states and even classify the fault nozzle among the 4 nozzles installed. The classification performance has a prediction accuracy of up to 99.2%. The result shows that the deep learning technique using melt pool images could be promising for in-situ DED process fault detection.

Vehicle abnormal jitter detection based on multi-task convolutional neural network

In order to overcome the problems of poor convergence, long detection time and high misjudgement rate existing in traditional methods, this paper proposes a vehicle abnormal jitter detection method based on multi-task convolutional neural network. The causes of abnormal vehicle jitter were analysed, and the equivalent taper of vehicles under different mileage was calculated to obtain the abnormal vehicle jitter frequency, and the time-frequency characteristics of vehicle vibration were obtained. The multi-task convolutional neural network is used to establish the calculation model of wheel matching geometric relationship under the condition of dithering, and the dithering degree of the vehicle body at different moments is obtained by sections, so as to realise the detection of abnormal vehicle dithering. The convergence coefficient of the method in this paper is higher than 0.9, the detection time of abnormal jitter is always below 1.2s, and the lowest misjudgement rate is only 0.6%.

Research on an Online Monitoring System for Efficient and Accurate Monitoring of Mine Water

Our goal is to solve the complexity of laying down wiring for traditional monitoring systems for mine water, while taking into account the poor timeliness of sampling and continuous monitoring, based on the processing and reuse of mine water. In this article, we focus on the theory of multi-sensor networks to estimate mine water quality, quantity monitoring equipment, and mine water Internet of Things (hereafter denoted “IoT”) communication systems. We designed a mine water IoT monitoring system for the wireless monitoring of water quantity and quality, developed an experimental platform for the wireless monitoring of mine water. This platform has the advantages of simple wiring and strong expandability. Finally, we tested and analyzed the system operation effects. To solve the problem of partial data abnormalities caused by problems of sensing equipment, we propose a data abnormality detection method based on an isolated forest that combines the characteristics of the sufficient timeliness of mine water monitoring data, and we offer experimental verification and analysis. The experimental results show that the system can realize the real-time wireless monitoring of mine water quality and quantity information with stable and fast data transmission capability. Furthermore, the system can quickly discover, analyze, and process abnormal data, has sufficient timeliness while guaranteeing the validity of the data output from the monitoring platform.

Abnormal detection method of accounting data based on information extraction technology

Abnormal detection method of accounting data based on information extraction technology

Abnormal detection method of accounting data based on information extraction technology

Abnormal detection method of accounting data based on information extraction technology