An automatic seismic signal detection method based on fourth-order statistics and applications

Real-time, automatic, and accurate determination of seismic signals is critical for rapid earthquake reporting and early warning. In this study, we present a correction trigger function（CTF） for automatically detecting regional seismic events and a fourth-order statistics algorithm with the Akaike information criterion（AIC） for determining the direct wave phase, based on the differences, or changes, in energy, frequency, and amplitude of the direct P- or S-waves signal and noise. Simulations suggest for that the proposed fourth-order statistics result in high resolution even for weak signal and noise variations at different amplitude, frequency, and polarization characteristics. To improve the precision of establishing the S-waves onset, first a specific segment of P-wave seismograms is selected and the polarization characteristics of the data are obtained. Second, the S-wave seismograms that contained the specific segment of P-wave seismograms are analyzed by S-wave polarization filtering. Finally, the S-wave phase onset times are estimated. The proposed algorithm was used to analyze regional earthquake data from the Shandong Seismic Network. The results suggest that compared with conventional methods, the proposed algorithm greatly decreased false and missed earthquake triggers, and improved the detection precision of direct P- and S-wave phases.

An Enhanced Lung Cancer Detection Approach Using Dual-Model Deep Learning Technique

Lung cancer continues to be a leading cause of cancer-related deaths worldwide,emphasizing the critical need for improved diagnostic techniques.Early detection of lung tumors significantly increases the chances of successful treatment and survival.However,current diagnostic methods often fail to detect tumors at an early stage or to accurately pinpoint their location within the lung tissue.Single-model deep learning technologies for lung cancer detection,while beneficial,cannot capture the full range of features present in medical imaging data,leading to incomplete or inaccurate detection.Furthermore,it may not be robust enough to handle the wide variability in medical images due to different imaging conditions,patient anatomy,and tumor characteristics.To overcome these disadvantages,dual-model or multi-model approaches can be employed.This research focuses on enhancing the detection of lung cancer by utilizing a combination of two learning models:a Convolutional Neural Network(CNN)for categorization and the You Only Look Once(YOLOv8)architecture for real-time identification and pinpointing of tumors.CNNs automatically learn to extract hierarchical features from raw image data,capturing patterns such as edges,textures,and complex structures that are crucial for identifying lung cancer.YOLOv8 incorporates multiscale feature extraction,enabling the detection of tumors of varying sizes and scales within a single image.This is particularly beneficial for identifying small or irregularly shaped tumors that may be challenging to detect.Furthermore,through the utilization of cutting-edge data augmentation methods,such as Deep Convolutional Generative Adversarial Networks(DCGAN),the suggested approach can handle the issue of limited data and boost the models’ability to learn from diverse and comprehensive datasets.The combined method not only improved accuracy and localization but also ensured efficient real-time processing,which is crucial for practical clinical applications.The CNN achieved an accuracy of 97.67%in classifying lung tissues into healthy and cancerous categories.The YOLOv8 model achieved an Intersection over Union(IoU)score of 0.85 for tumor localization,reflecting high precision in detecting and marking tumor boundaries within the images.Finally,the incorporation of synthetic images generated by DCGAN led to a 10%improvement in both the CNN classification accuracy and YOLOv8 detection performance.

MARIE:One-Stage Object Detection Mechanism for Real-Time Identifying of Firearms

Security and safety remain paramount concerns for both governments and individuals worldwide.In today’s context,the frequency of crimes and terrorist attacks is alarmingly increasing,becoming increasingly intolerable to society.Consequently,there is a pressing need for swift identification of potential threats to preemptively alert law enforcement and security forces,thereby preventing potential attacks or violent incidents.Recent advancements in big data analytics and deep learning have significantly enhanced the capabilities of computer vision in object detection,particularly in identifying firearms.This paper introduces a novel automatic firearm detection surveillance system,utilizing a one-stage detection approach named MARIE(Mechanism for Realtime Identification of Firearms).MARIE incorporates the Single Shot Multibox Detector(SSD)model,which has been specifically optimized to balance the speed-accuracy trade-off critical in firearm detection applications.The SSD model was further refined by integrating MobileNetV2 and InceptionV2 architectures for superior feature extraction capabilities.The experimental results demonstrate that this modified SSD configuration provides highly satisfactory performance,surpassing existing methods trained on the same dataset in terms of the critical speedaccuracy trade-off.Through these innovations,MARIE sets a new standard in surveillance technology,offering a robust solution to enhance public safety effectively.

Advancements in Liver Tumor Detection:A Comprehensive Review of Various Deep Learning Models

Liver cancer remains a leading cause of mortality worldwide,and precise diagnostic tools are essential for effective treatment planning.Liver Tumors(LTs)vary significantly in size,shape,and location,and can present with tissues of similar intensities,making automatically segmenting and classifying LTs from abdominal tomography images crucial and challenging.This review examines recent advancements in Liver Segmentation(LS)and Tumor Segmentation(TS)algorithms,highlighting their strengths and limitations regarding precision,automation,and resilience.Performance metrics are utilized to assess key detection algorithms and analytical methods,emphasizing their effectiveness and relevance in clinical contexts.The review also addresses ongoing challenges in liver tumor segmentation and identification,such as managing high variability in patient data and ensuring robustness across different imaging conditions.It suggests directions for future research,with insights into technological advancements that can enhance surgical planning and diagnostic accuracy by comparing popular methods.This paper contributes to a comprehensive understanding of current liver tumor detection techniques,provides a roadmap for future innovations,and improves diagnostic and therapeutic outcomes for liver cancer by integrating recent progress with remaining challenges.

Encrypted Cyberattack Detection System over Encrypted IoT Traffic Based onStatistical Intelligence

In the early days of IoT’s introduction, it was challenging to introduce encryption communication due to the lackof performance of each component, such as computing resources like CPUs and batteries, to encrypt and decryptdata. Because IoT is applied and utilized in many important fields, a cyberattack on IoT can result in astronomicalfinancial and human casualties. For this reason, the application of encrypted communication to IoT has beenrequired, and the application of encrypted communication to IoT has become possible due to improvements inthe computing performance of IoT devices and the development of lightweight cryptography. The applicationof encrypted communication in IoT has made it possible to use encrypted communication channels to launchcyberattacks. The approach of extracting evidence of an attack based on the primary information of a networkpacket is no longer valid because critical information, such as the payload in a network packet, is encrypted byencrypted communication. For this reason, technology that can detect cyberattacks over encrypted network trafficoccurring in IoT environments is required. Therefore, this research proposes an encrypted cyberattack detectionsystem for the IoT (ECDS-IoT) that derives valid features for cyberattack detection from the cryptographic networktraffic generated in the IoT environment and performs cyberattack detection based on the derived features. ECDS-IoT identifies identifiable information from encrypted traffic collected in IoT environments and extracts statistics-based features through statistical analysis of identifiable information. ECDS-IoT understands information aboutnormal data by learning only statistical features extracted from normal data. ECDS-IoT detects cyberattacks basedonly on the normal data information it has trained. To evaluate the cyberattack detection performance of theproposed ECDS-IoT in this research, ECDS-IoT used CICIoT2023, a dataset containing encrypted traffic generatedby normal and seven categories of cyberattacks in the IoT environment and experimented with cyberattackdetection on encrypted traffic using Autoencoder, RNN, GRU, LSTM, BiLSTM, and AE-LSTM algorithms. Asa result of evaluating the performance of cyberattack detection for encrypted traffic, ECDS-IoT achieved highperformance such as accuracy 0.99739, precision 0.99154, recall 1.0, F1 score 0.99575, and ROC_AUC 0.99822when using the AE-LSTM algorithm. As shown by the cyberattack detection results of ECDS-IoT, it is possibleto detect most cyberattacks through encrypted traffic. By applying ECDS-IoT to IoT, it can effectively detectcyberattacks concealed in encrypted traffic, promoting the efficient operation of IoT and preventing financial andhuman damage caused by cyberattacks.

Target detection for low angle radar based on multi-frequency order-statistics

For radar targets flying at low altitude, multiple pathways produce fade or enhancement relative to the level that would be expected in a free-space environment. In this paper, a new detec- tion method based on a wide-ranging multi-frequency radar for low angle targets is proposed. Sequential transmitting multiple pulses with different frequencies are first applied to decorrelate the cohe- rence of the direct and reflected echoes. After receiving all echoes, the multi-frequency samples are arranged in a sort descending ac- cording to the amplitude. Some high amplitude echoes in the same range cell are accumulated to improve the signal-to-noise ratio and the optimal number of high amplitude echoes is analyzed and given by experiments. Finally, simulation results are presented to verify the effectiveness of the method.

New Dynamic Algorithm for IRFPA Bad Pixel Detection and Compensation Based on Statistics

Based on the analysis to the behavior of bad pixels, a statistics-based auto-detecting and compensation algorithm for bad pixels is proposed. The correcting process is divided into two stages： bad pixel detection and bad pixel compensation. The proposed detection algorithm is a combination of median filtering and statistic method. Single frame median filtering is used to locate approximate map, then statistic method and threshold value is used to get the accurate location map of bad pixels. When the bad pixel detection is done, neighboring pixel replacement algorithm is used to compensate them in real-time. The effectiveness of this approach is test- ed by applying it to I-IgCATe infrared video. Experiments on real infrared imaging sequences demonstrate that the proposed algorithm requires only a few frames to obtain high quality corrections. It is easy to combine with traditional static methods, update the pre-defined location map in real-time.

Higher-Order Statistics for Automatic Weld Defect Detection

Image processing and image analysis are the main aspects for obtaining information from digital image owing to the fact that this techniques give the desired details in most of the applications generally and Non-Destructive testing specifically. This paper presents a proposed method for the automatic detection of weld defects in radiographic images. Firstly, the radiographic images were enhanced using adaptive histogram equalization and are filtered using mean and wiener filters. Secondly, the welding area is selected from the radiography image. Thirdly, the Cepstral features are extracted from the Higher-Order Spectra (Bispectrum and Trispectrum). Finally, neural networks are used for feature matching. The proposed method is tested using 100 radiographic images in the presence of noise and image blurring. Results show that in spite of time consumption, the proposed method yields best results for the automatic detection of weld defects in radiography images when the features were extracted from the Trispectrum of the image.

Speech-stream detection in short-wave channel based on empirical mode decomposition and higher-order statistics

To capture the presence of speech embedded in nonspeech events and background noise in shortwave non-cooperative communication, an algorithm for speech-stream detection in noisy environments is presented based on Empirical Mode Decomposition （EMD） and statistical properties of higher-order cumulants of speech signals. With the EMD, the noise signals can be decomposed into different numbers of IMFs. Then, the fourth-order cumulant （ FOC ） can be used to extract the desired feature of statistical properties for IMF components. Since the higher-order eumulants are blind for Gaussian signals, the proposed method is especially effective regarding the problem of speech-stream detection, where the speech signal is distorted by Gaussian noise. With the self-adaptive decomposition by EMD, the proposed method can also work well for non-Gaussian noise. The experiments show that the proposed algorithm can suppress different noise types with different SNRs, and the algorithm is robust in real signal tests.

An SDN Oriented Loop Detection Mechanism Based on TTL Statistics

The rise of the network has injected new impetus into the development of traditional networks.Due to the complexity of the network itself and its network programmability,there is a risk of routing loops occurring in the SDN network.This paper proposes a loop detection mechanism.According to the Time To Live(TTL)value of the loop packet,there is approximately periodicity in the same loop.We use sFlow to count the number of packets corresponding to each TTL value of a switch in the loop over a period of time,and perform discrete Fourier transform on the obtained finite-length sequence to observe its frequency domain performance and determine whether there are periodic features.By doing so,it is determined whether there is a routing loop,and the purpose of passively detecting the routing loop is achieved.Compared to existing algorithms,it has advantages in real-time,scalability and false positive rate.The experimental results show that the routing loop detection algorithm based on TTL statistics in this paper still maintains high judgment accuracy under the scenarios of lower stream sampling rate and smaller detection period.

Auroral event detection using spatiotemporal statistics of local motion vectors

The analysis and exploration of auroral dynamics are very significant for studying auroral mechanisms. This paper proposes a method based on auroral dynamic processes for detecting auroral events automatically. We first obtained the motion fields using the multiscale fluid flow estimator. Then, the auroral video frame sequence was represented by the spatiotemporal statistics of local motion vectors. Finally, automatic auroral event detection was achieved. The experimental results show that our methods could detect the required auroral events effectively and accurately, and that the detections were independent on any specific auroral event. The proposed method makes it feasible to statistically analyze a large number of continuous observations based on the auroral dynamic process.

Adaptive Noise Detection and Removal Algorithm Using Local Statistics for Salt-and-Pepper Noise

In this paper, an adaptive noise detection and removal algorithm using local statistics for salt-and-pepper noise are proposed. In order to determine constraints for noise detection, the local mean, varianoe, and maximum value are used. In addition, a weighted median filter is employed to remove the detected noise. The simulation results show the capability of the proposed algorithm removes the noise effectively.

Statistics Modeling of Shallow Sea Ambient Noise and Its Applications in Low-frequency Line Spectrum Detection

The noise's statistical characteristics are very important for signal detection.In this paper,the ambient noise statistical characteristics are investigated by using the recorded noise data in sea trials first,and the results show that the generalized Gaussian distribution is a suitable model for the ambient noise modeling.Thereafter,the optimal detector based on maximum likelihood ratio can be deduced,and the asymptotic detector is also derived under weak signal assumption.The detector's performance is verified by using numerical simulation,and the results showthat the optimal and asymptotic detectors outperform the conventional correlation-integration system due to accuracy modeling of ambient noise.

Comparison and Adaptation of Two Strategies for Anomaly Detection in Load Profiles Based on Methods from the Fields of Machine Learning and Statistics

The Federal Office for Economic Affairs and Export Control (BAFA) of Germany promotes digital concepts for increasing energy efficiency as part of the “Pilotprogramm Einsparzähler”. Within this program, Limón GmbH is developing software solutions in cooperation with the University of Kassel to identify efficiency potentials in load profiles by means of automated anomaly detection. Therefore, in this study two strategies for anomaly detection in load profiles are evaluated. To estimate the monthly load profile, strategy 1 uses the artificial neural network LSTM (Long Short-Term Memory), with a data period of one month (1 M) or three months (3 M), and strategy 2 uses the smoothing method PEWMA (Probalistic Exponential Weighted Moving Average). By comparing with original load profile data, residuals or summed residuals of the sequence lengths of two, four, six and eight hours are identified as an anomaly by exceeding a predefined threshold. The thresholds are defined by the Z-Score test, i.e., residuals greater than 2, 2.5 or 3 standard deviations are considered anomalous. Furthermore, the ESD (Extreme Studentized Deviate) test is used to set thresholds by means of three significance level values of 0.05, 0.10 and 0.15, with a maximum of k = 40 iterations. Five load profiles are examined, which were obtained by the cluster method k-Means as a representative sample from all available data sets of the Limón GmbH. The evaluation shows that for strategy 1 a maximum F1-value of 0.4 (1 M) and for all examined companies an average F1-value of maximum 0.24 and standard deviation of 0.09 (1 M) could be achieved for the investigation on single residuals. In variant 3 M the highest F1-value could be achieved with an average F1-value of 0.21 and standard deviation of 0.06 (3 M) for summed residuals of the partial sequence length of four hours. The PEWMA-based strategy 2 did not show a higher anomaly detection efficacy compared to strategy 1 in any of the investigated companies.

基于改进Detection Transformer的棉花幼苗与杂草检测模型研究

基于深度学习的目标检测技术在棉花幼苗与杂草检测领域已取得一定进展。论文提出了基于改进Detection Transformer的棉花幼苗与杂草检测模型,以提高杂草目标检测的准确率和效率。首先,引入了可变形注意力模块替代原始模型中的Transformer注意力模块,提高模型对特征图目标形变的处理能力。提出新的降噪训练机制,解决了二分图匹配不稳定问题。提出混合查询选择策略,提高解码器对目标类别和位置信息的利用效率。使用Swin Transformer作为网络主干,提高模型特征提取能力。通过对比原网络,论文提出的模型方法在训练过程中表现出更快的收敛速度,并且在准确率方面提高了6.7%。

Improvement of High-Speed Detection Algorithm for Nonwoven Material Defects Based on Machine Vision

Defect detection is vital in the nonwoven material industry,ensuring surface quality before producing finished products.Recently,deep learning and computer vision advancements have revolutionized defect detection,making it a widely adopted approach in various industrial fields.This paper mainly studied the defect detection method for nonwoven materials based on the improved Nano Det-Plus model.Using the constructed samples of defects in nonwoven materials as the research objects,transfer learning experiments were conducted based on the Nano DetPlus object detection framework.Within this framework,the Backbone,path aggregation feature pyramid network(PAFPN)and Head network models were compared and trained through a process of freezing,with the ultimate aim of bolstering the model's feature extraction abilities and elevating detection accuracy.The half-precision quantization method was used to optimize the model after transfer learning experiments,reducing model weights and computational complexity to improve the detection speed.Performance comparisons were conducted between the improved model and the original Nano Det-Plus model,YOLO,SSD and other common industrial defect detection algorithms,validating that the improved methods based on transfer learning and semi-precision quantization enabled the model to meet the practical requirements of industrial production.

A Hybrid Intrusion Detection Method Based on Convolutional Neural Network and AdaBoost

To solve the problem of poor detection and limited application range of current intrusion detection methods,this paper attempts to use deep learning neural network technology to study a new type of intrusion detection method.Hence,we proposed an intrusion detection algorithm based on convolutional neural network(CNN)and AdaBoost algorithm.This algorithm uses CNN to extract the characteristics of network traffic data,which is particularly suitable for the analysis of continuous and classified attack data.The AdaBoost algorithm is used to classify network attack data that improved the detection effect of unbalanced data classification.We adopt the UNSW-NB15 dataset to test of this algorithm in the PyCharm environment.The results show that the detection rate of algorithm is99.27%and the false positive rate is lower than 0.98%.Comparative analysis shows that this algorithm has advantages over existing methods in terms of detection rate and false positive rate for small proportion of attack data.

Statistical moment-based structural damage detection method in time domain

A novel structural damage detection method with a new damage index,i.e.,the statistical moment-based damage detection（SMBDD） method in the frequency domain,has been recently proposed.The aim of this study is to extend the SMBDD method in the frequency domain to the time domain for building structures subjected to non-Gaussian and non-stationary excitations.The applicability and effectiveness of the SMBDD method in the time domainis verified both numerically and experimentally.Shear buildings with various damage scenarios are first numerically investigated in the time domain taking into account the effect of measurement noise.The applicability of the proposed method in the time domain to building structures subjected to non-Gaussian and non-stationary excitations is then experimentally investigated through a series of shaking table tests,in which two three-story shear building models with four damage scenarios aretested.The identified damage locations and severities are then compared with the preset values.The comparative results are found to be satisfactory,and the SMBDD method is shown to be feasible and effective for building structures subjected to non-Gaussian and non-stationary excitations.

IDS-INT:Intrusion detection system using transformer-based transfer learning for imbalanced network traffic

A network intrusion detection system is critical for cyber security against llegitimate attacks.In terms of feature perspectives,network traffic may include a variety of elements such as attack reference,attack type,a subcategory of attack,host information,malicious scripts,etc.In terms of network perspectives,network traffic may contain an imbalanced number of harmful attacks when compared to normal traffic.It is challenging to identify a specific attack due to complex features and data imbalance issues.To address these issues,this paper proposes an Intrusion Detection System using transformer-based transfer learning for Imbalanced Network Traffic(IDS-INT).IDS-INT uses transformer-based transfer learning to learn feature interactions in both network feature representation and imbalanced data.First,detailed information about each type of attack is gathered from network interaction descriptions,which include network nodes,attack type,reference,host information,etc.Second,the transformer-based transfer learning approach is developed to learn detailed feature representation using their semantic anchors.Third,the Synthetic Minority Oversampling Technique(SMOTE)is implemented to balance abnormal traffic and detect minority attacks.Fourth,the Convolution Neural Network(CNN)model is designed to extract deep features from the balanced network traffic.Finally,the hybrid approach of the CNN-Long Short-Term Memory(CNN-LSTM)model is developed to detect different types of attacks from the deep features.Detailed experiments are conducted to test the proposed approach using three standard datasets,i.e.,UNsWNB15,CIC-IDS2017,and NSL-KDD.An explainable AI approach is implemented to interpret the proposed method and develop a trustable model.

Automated Vulnerability Detection of Blockchain Smart Contacts Based on BERT Artificial Intelligent Model

The widespread adoption of blockchain technology has led to the exploration of its numerous applications in various fields.Cryptographic algorithms and smart contracts are critical components of blockchain security.Despite the benefits of virtual currency,vulnerabilities in smart contracts have resulted in substantial losses to users.While researchers have identified these vulnerabilities and developed tools for detecting them,the accuracy of these tools is still far from satisfactory,with high false positive and false negative rates.In this paper,we propose a new method for detecting vulnerabilities in smart contracts using the BERT pre-training model,which can quickly and effectively process and detect smart contracts.More specifically,we preprocess and make symbol substitution in the contract,which can make the pre-training model better obtain contract features.We evaluate our method on four datasets and compare its performance with other deep learning models and vulnerability detection tools,demonstrating its superior accuracy.