Comparative Analysis of ARIMA and LSTM Model-Based Anomaly Detection for Unannotated Structural Health Monitoring Data in an Immersed Tunnel

Structural Health Monitoring(SHM)systems have become a crucial tool for the operational management of long tunnels.For immersed tunnels exposed to both traffic loads and the effects of the marine environment,efficiently identifying abnormal conditions from the extensive unannotated SHM data presents a significant challenge.This study proposed amodel-based approach for anomaly detection and conducted validation and comparative analysis of two distinct temporal predictive models using SHM data from a real immersed tunnel.Firstly,a dynamic predictive model-based anomaly detectionmethod is proposed,which utilizes a rolling time window for modeling to achieve dynamic prediction.Leveraging the assumption of temporal data similarity,an interval prediction value deviation was employed to determine the abnormality of the data.Subsequently,dynamic predictive models were constructed based on the Autoregressive Integrated Moving Average(ARIMA)and Long Short-Term Memory(LSTM)models.The hyperparameters of these models were optimized and selected using monitoring data from the immersed tunnel,yielding viable static and dynamic predictive models.Finally,the models were applied within the same segment of SHM data,to validate the effectiveness of the anomaly detection approach based on dynamic predictive modeling.A detailed comparative analysis discusses the discrepancies in temporal anomaly detection between the ARIMA-and LSTM-based models.The results demonstrated that the dynamic predictive modelbased anomaly detection approach was effective for dealing with unannotated SHM data.In a comparison between ARIMA and LSTM,it was found that ARIMA demonstrated higher modeling efficiency,rendering it suitable for short-term predictions.In contrast,the LSTM model exhibited greater capacity to capture long-term performance trends and enhanced early warning capabilities,thereby resulting in superior overall performance.

Finite element model updating for structural damage detection using transmissibility data

This paper presents a new finite element model updating method for estimating structural parameters and detecting structural damage location and severity based on the structural responses(output-only data).The method uses the sensitivity relation of transmissibility data through a least-squares algorithm and appropriate normalization of the extracted equations.The proposed transmissibility-based sensitivity equation produces a more significant number of equations than the sensitivity equations based on the frequency response function(FRF),which can estimate the structural parameters with higher accuracy.The abilities of the proposed method are assessed by using numerical data of a two-story two-bay frame model and a plate structure model.In evaluating different damage cases,the number,location,and stiffness reduction of the damaged elements and the severity of the simulated damage have been accurately identified.The reliability and stability of the presented method against measurement and modeling errors are examined using error-contaminated data.The parameter estimation results prove the method’s capabilities as an accurate model updating algorithm.

Enhancing PDF Malware Detection through Logistic Model Trees

Malware is an ever-present and dynamic threat to networks and computer systems in cybersecurity,and because of its complexity and evasiveness,it is challenging to identify using traditional signature-based detection approaches.The study article discusses the growing danger to cybersecurity that malware hidden in PDF files poses,highlighting the shortcomings of conventional detection techniques and the difficulties presented by adversarial methodologies.The article presents a new method that improves PDF virus detection by using document analysis and a Logistic Model Tree.Using a dataset from the Canadian Institute for Cybersecurity,a comparative analysis is carried out with well-known machine learning models,such as Credal Decision Tree,Naïve Bayes,Average One Dependency Estimator,Locally Weighted Learning,and Stochastic Gradient Descent.Beyond traditional structural and JavaScript-centric PDF analysis,the research makes a substantial contribution to the area by boosting precision and resilience in malware detection.The use of Logistic Model Tree,a thorough feature selection approach,and increased focus on PDF file attributes all contribute to the efficiency of PDF virus detection.The paper emphasizes Logistic Model Tree’s critical role in tackling increasing cybersecurity threats and proposes a viable answer to practical issues in the sector.The results reveal that the Logistic Model Tree is superior,with improved accuracy of 97.46%when compared to benchmark models,demonstrating its usefulness in addressing the ever-changing threat landscape.

DCEL:classifier fusion model for Android malware detection

The rapid growth of mobile applications,the popularity of the Android system and its openness have attracted many hackers and even criminals,who are creating lots of Android malware.However,the current methods of Android malware detection need a lot of time in the feature engineering phase.Furthermore,these models have the defects of low detection rate,high complexity,and poor practicability,etc.We analyze the Android malware samples,and the distribution of malware and benign software in application programming interface(API)calls,permissions,and other attributes.We classify the software’s threat levels based on the correlation of features.Then,we propose deep neural networks and convolutional neural networks with ensemble learning(DCEL),a new classifier fusion model for Android malware detection.First,DCEL preprocesses the malware data to remove redundant data,and converts the one-dimensional data into a two-dimensional gray image.Then,the ensemble learning approach is used to combine the deep neural network with the convolutional neural network,and the final classification results are obtained by voting on the prediction of each single classifier.Experiments based on the Drebin and Malgenome datasets show that compared with current state-of-art models,the proposed DCEL has a higher detection rate,higher recall rate,and lower computational cost.

Model Agnostic Meta-Learning(MAML)-Based Ensemble Model for Accurate Detection of Wheat Diseases Using Vision Transformer and Graph Neural Networks

Wheat is a critical crop,extensively consumed worldwide,and its production enhancement is essential to meet escalating demand.The presence of diseases like stem rust,leaf rust,yellow rust,and tan spot significantly diminishes wheat yield,making the early and precise identification of these diseases vital for effective disease management.With advancements in deep learning algorithms,researchers have proposed many methods for the automated detection of disease pathogens;however,accurately detectingmultiple disease pathogens simultaneously remains a challenge.This challenge arises due to the scarcity of RGB images for multiple diseases,class imbalance in existing public datasets,and the difficulty in extracting features that discriminate between multiple classes of disease pathogens.In this research,a novel method is proposed based on Transfer Generative Adversarial Networks for augmenting existing data,thereby overcoming the problems of class imbalance and data scarcity.This study proposes a customized architecture of Vision Transformers(ViT),where the feature vector is obtained by concatenating features extracted from the custom ViT and Graph Neural Networks.This paper also proposes a Model AgnosticMeta Learning(MAML)based ensemble classifier for accurate classification.The proposedmodel,validated on public datasets for wheat disease pathogen classification,achieved a test accuracy of 99.20%and an F1-score of 97.95%.Compared with existing state-of-the-art methods,this proposed model outperforms in terms of accuracy,F1-score,and the number of disease pathogens detection.In future,more diseases can be included for detection along with some other modalities like pests and weed.

Intrusion Detection Model Using Chaotic MAP for Network Coding Enabled Mobile Small Cells

Wireless Network security management is difficult because of the ever-increasing number of wireless network malfunctions,vulnerabilities,and assaults.Complex security systems,such as Intrusion Detection Systems(IDS),are essential due to the limitations of simpler security measures,such as cryptography and firewalls.Due to their compact nature and low energy reserves,wireless networks present a significant challenge for security procedures.The features of small cells can cause threats to the network.Network Coding(NC)enabled small cells are vulnerable to various types of attacks.Avoiding attacks and performing secure“peer”to“peer”data transmission is a challenging task in small cells.Due to the low power and memory requirements of the proposed model,it is well suited to use with constrained small cells.An attacker cannot change the contents of data and generate a new Hashed Homomorphic Message Authentication Code(HHMAC)hash between transmissions since the HMAC function is generated using the shared secret.In this research,a chaotic sequence mapping based low overhead 1D Improved Logistic Map is used to secure“peer”to“peer”data transmission model using lightweight H-MAC(1D-LM-P2P-LHHMAC)is proposed with accurate intrusion detection.The proposed model is evaluated with the traditional models by considering various evaluation metrics like Vector Set Generation Accuracy Levels,Key Pair Generation Time Levels,Chaotic Map Accuracy Levels,Intrusion Detection Accuracy Levels,and the results represent that the proposed model performance in chaotic map accuracy level is 98%and intrusion detection is 98.2%.The proposed model is compared with the traditional models and the results represent that the proposed model secure data transmission levels are high.

Similarity evaluation model for the internal defect detection of strip steel

An internal defect meter is an instrument to detect the internal inclusion defects of cold-rolled strip steel.The detection accuracy of the equipment can be evaluated based on the similarity of the multiple detection data obtained for the same steel coil.Based on the cosine similarity model and eigenvalue matrix model,a comprehensive evaluation method to calculate the weighted average of similarity is proposed.Results show that the new method is consistent with and can even replace artificial evaluation to realize the automatic evaluation of strip defect detection results.

Surface Defect Detection and Evaluation Method of Large Wind Turbine Blades Based on an Improved Deeplabv3+Deep Learning Model

The accumulation of defects on wind turbine blade surfaces can lead to irreversible damage,impacting the aero-dynamic performance of the blades.To address the challenge of detecting and quantifying surface defects on wind turbine blades,a blade surface defect detection and quantification method based on an improved Deeplabv3+deep learning model is proposed.Firstly,an improved method for wind turbine blade surface defect detection,utilizing Mobilenetv2 as the backbone feature extraction network,is proposed based on an original Deeplabv3+deep learning model to address the issue of limited robustness.Secondly,through integrating the concept of pre-trained weights from transfer learning and implementing a freeze training strategy,significant improvements have been made to enhance both the training speed and model training accuracy of this deep learning model.Finally,based on segmented blade surface defect images,a method for quantifying blade defects is proposed.This method combines image stitching algorithms to achieve overall quantification and risk assessment of the entire blade.Test results show that the improved Deeplabv3+deep learning model reduces training time by approximately 43.03%compared to the original model,while achieving mAP and MIoU values of 96.87%and 96.93%,respectively.Moreover,it demonstrates robustness in detecting different surface defects on blades across different back-grounds.The application of a blade surface defect quantification method enables the precise quantification of dif-ferent defects and facilitates the assessment of risk levels associated with defect measurements across the entire blade.This method enables non-contact,long-distance,high-precision detection and quantification of surface defects on the blades,providing a reference for assessing surface defects on wind turbine blades.

Research on Track Fastener Service Status Detection Based on Improved Yolov4 Model

As an important part of railway lines, the healthy service status of track fasteners was very important to ensure the safety of trains. The application of deep learning algorithms was becoming an important method to realize its state detection. However, there was often a deficiency that the detection accuracy and calculation speed of model were difficult to balance, when the traditional deep learning model is used to detect the service state of track fasteners. Targeting this issue, an improved Yolov4 model for detecting the service status of track fasteners was proposed. Firstly, the Mixup data augmentation technology was introduced into Yolov4 model to enhance the generalization ability of model. Secondly, the MobileNet-V2 lightweight network was employed in lieu of the CSPDarknet53 network as the backbone, thereby reducing the number of algorithm parameters and improving the model’s computational efficiency. Finally, the SE attention mechanism was incorporated to boost the importance of rail fastener identification by emphasizing relevant image features, ensuring that the network’s focus was primarily on the fasteners being inspected. The algorithm achieved both high precision and high speed operation of the rail fastener service state detection, while realizing the lightweight of model. The experimental results revealed that, the MAP value of the rail fastener service state detection algorithm based on the improved Yolov4 model reaches 83.2%, which is 2.83% higher than that of the traditional Yolov4 model, and the calculation speed was improved by 67.39%. Compared with the traditional Yolov4 model, the proposed method achieved the collaborative optimization of detection accuracy and calculation speed.

A Double-Interactively Recurrent Fuzzy Cerebellar Model Articulation Controller Model Combined with an Improved Particle Swarm Optimization Method for Fall Detection

In many Eastern and Western countries,falling birth rates have led to the gradual aging of society.Older adults are often left alone at home or live in a long-term care center,which results in them being susceptible to unsafe events(such as falls)that can have disastrous consequences.However,automatically detecting falls fromvideo data is challenging,and automatic fall detection methods usually require large volumes of training data,which can be difficult to acquire.To address this problem,video kinematic data can be used as training data,thereby avoiding the requirement of creating a large fall data set.This study integrated an improved particle swarm optimization method into a double interactively recurrent fuzzy cerebellar model articulation controller model to develop a costeffective and accurate fall detection system.First,it obtained an optical flow(OF)trajectory diagram from image sequences by using the OF method,and it solved problems related to focal length and object offset by employing the discrete Fourier transform(DFT)algorithm.Second,this study developed the D-IRFCMAC model,which combines spatial and temporal(recurrent)information.Third,it designed an IPSO(Improved Particle Swarm Optimization)algorithm that effectively strengthens the exploratory capabilities of the proposed D-IRFCMAC(Double-Interactively Recurrent Fuzzy Cerebellar Model Articulation Controller)model in the global search space.The proposed approach outperforms existing state-of-the-art methods in terms of action recognition accuracy on the UR-Fall,UP-Fall,and PRECIS HAR data sets.The UCF11 dataset had an average accuracy of 93.13%,whereas the UCF101 dataset had an average accuracy of 92.19%.The UR-Fall dataset had an accuracy of 100%,the UP-Fall dataset had an accuracy of 99.25%,and the PRECIS HAR dataset had an accuracy of 99.07%.

Rock physics modeling of heterogeneous carbonatereservoirs： porosity estimation and hydrocarbon detection

In heterogeneous natural gas reservoirs, gas is generally present as small patchlike pockets embedded in the water-saturated host matrix. This type of heterogeneity, also called "patchy saturation", causes significant seismic velocity dispersion and attenuation. To establish the relation between seismic response and type of fluids, we designed a rock physics model for carbonates. First, we performed CT scanning and analysis of the fluid distribution in the partially saturated rocks. Then, we predicted the quantitative relation between the wave response at different frequency ranges and the basic lithological properties and pore fluids. A rock physics template was constructed based on thin section analysis of pore structures and seismic inversion. This approach was applied to the limestone gas reservoirs of the right bank block of the Amu Darya River. Based on poststack wave impedance and prestack elastic parameter inversions, the seismic data were used to estimate rock porosity and gas saturation. The model results were in good agreement with the production regime of the wells.

The analysis of frequency-dependent characteristics for fluid detection: a physical model experiment

According to the Chapman multi-scale rock physical model, the seismic response characteristics vary for different fluid-saturated reservoirs. For class I AVO reservoirs and gas-saturation, the seismic response is a high-frequency bright spot as the amplitude energy shifts. However, it is a low-frequency shadow for the Class III AVO reservoirs saturated with hydrocarbons. In this paper, we verified the high-frequency bright spot results of Chapman for the Class I AVO response using the frequency-dependent analysis of a physical model dataset. The physical model is designed as inter-bedded thin sand and shale based on real field geology parameters. We observed two datasets using fixed offset and 2D geometry with different fluid- saturated conditions. Spectral and time-frequency analyses methods are applied to the seismic datasets to describe the response characteristics for gas-, water-, and oil-saturation. The results of physical model dataset processing and analysis indicate that reflection wave tuning and fluid-related dispersion are the main seismic response characteristic mechanisms. Additionally, the gas saturation model can be distinguished from water and oil saturation for Class I AVO utilizing the frequency-dependent abnormal characteristic. The frequency-dependent characteristic analysis of the physical model dataset verified the different spectral response characteristics corresponding to the different fluid-saturated models. Therefore, by careful analysis of real field seismic data, we can obtain the abnormal spectral characteristics induced by the fluid variation and implement fluid detection using seismic data directly.

Adaptive moving target detection algorithm based on Gaussian mixture model

In order to enhance the reliability of the moving target detection, an adaptive moving target detection algorithm based on the Gaussian mixture model is proposed. This algorithm employs Gaussian mixture distributions in modeling the background of each pixel. As a result, the number of Gaussian distributions is not fixed but adaptively changes with the change of the pixel value frequency. The pixels of the difference image are divided into two parts according to their values. Then the two parts are separately segmented by the adaptive threshold, and finally the foreground image is obtained. The shadow elimination method based on morphological reconstruction is introduced to improve the performance of foreground image＇s segmentation. Experimental results show that the proposed algorithm can quickly and accurately build the background model and it is more robust in different real scenes.

Customized Convolutional Neural Network for Accurate Detection of Deep Fake Images in Video Collections

The motivation for this study is that the quality of deep fakes is constantly improving,which leads to the need to develop new methods for their detection.The proposed Customized Convolutional Neural Network method involves extracting structured data from video frames using facial landmark detection,which is then used as input to the CNN.The customized Convolutional Neural Network method is the date augmented-based CNN model to generate‘fake data’or‘fake images’.This study was carried out using Python and its libraries.We used 242 films from the dataset gathered by the Deep Fake Detection Challenge,of which 199 were made up and the remaining 53 were real.Ten seconds were allotted for each video.There were 318 videos used in all,199 of which were fake and 119 of which were real.Our proposedmethod achieved a testing accuracy of 91.47%,loss of 0.342,and AUC score of 0.92,outperforming two alternative approaches,CNN and MLP-CNN.Furthermore,our method succeeded in greater accuracy than contemporary models such as XceptionNet,Meso-4,EfficientNet-BO,MesoInception-4,VGG-16,and DST-Net.The novelty of this investigation is the development of a new Convolutional Neural Network(CNN)learning model that can accurately detect deep fake face photos.

Efficient Deviation Detection Between a Process Model and Event Logs

Business processes described by formal or semi-formal models are realized via information systems.Event logs generated from these systems are probably not consistent with the existing models due to insufficient design of the information system or the system upgrade.By comparing an existing process model with event logs,we can detect inconsistencies called deviations,verify and extend the business process model,and accordingly improve the business process.In this paper,some abnormal activities in business processes are formally defined based on Petri nets.An efficient approach to detect deviations between the process model and event logs is proposed.Then,business process models are revised when abnormal activities exist.A clinical process in a healthcare information system is used as a case study to illustrate our work.Experimental results show the effectiveness and efficiency of the proposed approach.

Intrusion Detection Algorithm Based on Model Checking Interval Temporal Logic

Model checking based on linear temporal logic reduces the false negative rate of misuse detection.However,linear temporal logic formulae cannot be used to describe concurrent attacks and piecewise attacks.So there is still a high rate of false negatives in detecting these complex attack patterns.To solve this problem,we use interval temporal logic formulae to describe concurrent attacks and piecewise attacks.On this basis,we formalize a novel algorithm for intrusion detection based on model checking interval temporal logic.Compared with the method based on model checking linear temporal logic,the new algorithm can find unknown succinct attacks.The simulation results show that the new method can effectively reduce the false negative rate of concurrent attacks and piecewise attacks.

Quantitative Damage Detection for Planetary Gear Sets Based on Physical Models

Planetary gear set is the critical component in helicopter transmission train, and an important problem in condition monitoring and health management of planetary gear set is quantitative damage detection. In order to resolve this problem, an approach based on physical models is presented to detect damage quantitatively in planetary gear set. A particular emphasis is put on a feature generation and selection method, which is used for sun gear tooth breakage damage detection quantitatively in planetary gear box of helicopter transmission system. In this feature generation procedure, the pure torsional dynamical models of 2K-H planetary gear set is established for healthy case and sun gear tooth-breakage case. Then, a feature based on the spectrum of simulation signals of the dynamical models is generated. Aiming at selecting the best feature suitable for quantitative damage detection, a two-sample Z-test procedure is used to analyze the performance of features on damage evolution tracing. A feature named SR, which had better performance in tracking damage, is proposed to detect damage in planetary gear set. Meanwhile, the sun gear tooth-chipped seeded experiments with different severity are designed to validate the method above, and then the test vibration signal is picked up and used for damage detection. With the results of several experiments for quantitative damage detection, the feasibility and the effect of this approach are verified. The proposed method can supply an effective tool for degradation state identification in condition monitoring and health management of helicopter transmission system.

Grey-theory based intrusion detection model

To solve the problem that current intrusion detection model needs large-scale data in formulating the model in real-time use, an intrusion detection system model based on grey theory （GTIDS） is presented. Grey theory has merits of fewer requirements on original data scale, less limitation of the distribution pattern and simpler algorithm in modeling. With these merits GTIDS constructs model according to partial time sequence for rapid detect on intrusive act in secure system. In this detection model rate of false drop and false retrieval are effectively reduced through twice modeling and repeated detect on target data. Furthermore, GTIDS framework and specific process of modeling algorithm are presented. The affectivity of GTIDS is proved through emulated experiments comparing snort and next-generation intrusion detection expert system （NIDES） in SRI international.

A Classification–Detection Approach of COVID-19 Based on Chest X-ray and CT by Using Keras Pre-Trained Deep Learning Models

The Coronavirus Disease 2019(COVID-19)is wreaking havoc around the world,bring out that the enormous pressure on national health and medical staff systems.One of the most effective and critical steps in the fight against COVID-19,is to examine the patient’s lungs based on the Chest X-ray and CT generated by radiation imaging.In this paper,five keras-related deep learning models:ResNet50,InceptionResNetV2,Xception,transfer learning and pre-trained VGGNet16 is applied to formulate an classification-detection approaches of COVID-19.Two benchmark methods SVM(Support Vector Machine),CNN(Conventional Neural Networks)are provided to compare with the classification-detection approaches based on the performance indicators,i.e.,precision,recall,F1 scores,confusion matrix,classification accuracy and three types of AUC(Area Under Curve).The highest classification accuracy derived by classification-detection based on 5857 Chest X-rays and 767 Chest CTs are respectively 84%and 75%,which shows that the keras-related deep learning approaches facilitate accurate and effective COVID-19-assisted detection.

An Intrusion Detection Method Based on Hierarchical Hidden Markov Models

This paper presents an anomaly detection approach to detect intrusions into computer systems. In this approach, a hierarchical hidden Markov model （HHMM） is used to represent a temporal profile of normal behavior in a computer system. The HHMM of the norm profile is learned from historic data of the system＇s normal behavior. The observed behavior of the system is analyzed to infer the probability that the HHMM of the norm profile supports the observed behavior. A low probability of support indicates an anomalous behavior that may result from intrusive activities. The model was implemented and tested on the UNIX system call sequences collected by the University of New Mexico group. The testing results showed that the model can clearly identify the anomaly activities and has a better performance than hidden Markov model.