Real-time Detection Method Research Articles

Machine Learning Assisted-Intelligent Lactic Acid Monitoring in Sweat Supported by a Perspiration-Driven Self-Powered Sensor.

Lactic acid has aroused increasing attention due to its close association with serious diseases. A real-time, dynamic, and intelligent detection method is vital for sensitive detection of lactic acid. Here, a machine learning (ML)-assisted perspiration-driven self-powered sensor (PDS sensor) is fabricated using Ni-ZIF-8@lactate oxidase and pyruvate oxidase (Ni-ZIF-8@LOx&POx)/laser-induced graphene (LIG), bilirubin oxidase (BOD)/LIG, and a microchannel for highly sensitive and real-time monitoring of lactic acid in sweat. Driven by the oxidation reaction of lactic acid, PDS sensors exhibit excellent sensitivity, a wide detection range, good reproducibility, and excellent selectivity for lactic acid detection in sweat. When subjects with different body mass index (BMI) undergo aerobic or anaerobic exercise or maintain a sedentary state, PDS sensors can monitor lactic acid in sweat wirelessly and in real-time. Moreover, a ML algorithm was employed to assist PDS sensors to detect lactic acid in the subjects' sweat with a high prediction accuracy of 96.0%.

Real-time detection method of intelligent classification and defect of transmission line insulator based on LightWeight-YOLOv8n network

Real-time detection method of intelligent classification and defect of transmission line insulator based on LightWeight-YOLOv8n network

Binocular Video-Based Automatic Pixel-Level Crack Detection and Quantification Using Deep Convolutional Neural Networks for Concrete Structures

Crack detection and quantification play crucial roles in assessing the condition of concrete structures. Herein, a novel real-time crack detection and quantification method that leverages binocular vision and a lightweight deep learning model is proposed. In this methodology, the proposed method based on the following four modules is adopted: a lightweight classification algorithm, a high-precision segmentation algorithm, a semi-global block matching algorithm (SGBM), and a crack quantification technique. Based on the crack segmentation results, a framework is developed for quantitative analysis of the major geometric parameters, including crack length, crack width, and crack angle of orientation at the pixel level. Results indicate that, by incorporating channel attention and spatial attention mechanisms in the MBConv module, the detection accuracy of the improved EfficientNetV2 increased by 1.6% compared with the original EfficientNetV2. Results indicate that using the proposed quantification method can achieve low quantification errors of 2%, 4.5%, and 4% for the crack length, width, and angle of orientation, respectively. The proposed method can contribute to crack detection and quantification in practical use by being deployed on smart devices.

Cardioattentionnet: advancing ECG beat characterization with a high-accuracy and portable deep learning model.

The risk of mortality associated with cardiac arrhythmias is considerable, and their diagnosis presents significant challenges, often resulting in misdiagnosis. This situation highlights the necessity for an automated, efficient, and real-time detection method aimed at enhancing diagnostic accuracy and improving patient outcomes. The present study is centered on the development of a portable deep learning model for the detection of arrhythmias via electrocardiogram (ECG) signals, referred to as CardioAttentionNet (CANet). CANet integrates Bi-directional Long Short-Term Memory (BiLSTM) networks, Multi-head Attention mechanisms, and Depthwise Separable Convolution, thereby facilitating its application in portable devices for early diagnosis. The architecture of CANet allows for effective processing of extended ECG patterns and detailed feature extraction without a substantial increase in model size. Empirical results indicate that CANet outperformed traditional models in terms of predictive performance and stability, as confirmed by comprehensive cross-validation. The model demonstrated exceptional capabilities in detecting cardiac arrhythmias, surpassing existing models in both cross-validation and external testing scenarios. Specifically, CANet achieved high accuracy in classifying various arrhythmic events, with the following accuracies reported for different categories: Normal (97.37 ± 0.30%), Supraventricular (98.09 ± 0.25%), Ventricular (92.92 ± 0.09%), Atrial Fibrillation (99.07 ± 0.13%), and Unclassified arrhythmias (99.68 ± 0.06%). In external evaluations, CANet attained an average accuracy of 94.41%, with the area under the curve (AUC) for each category exceeding 99%, thereby demonstrating its substantial clinical applicability and significant advancements over traditional models. The deep learning model proposed in this study has the potential to enhance the accuracy of early diagnosis for various types of arrhythmias. Looking ahead, this technology is anticipated to provide improved medical services for patients with heart disease through continuous, non-invasive monitoring and timely intervention.

Direct mathematical method for real-time ischemic episodes detection from electrocardiograms using the discrete Hermite transform

A real-time automated identification technique is developed for the detection of ischemic episodes in long-term electrocardiographic (ECG) signals using mathematical expansions involving the Discrete Dilated Hermite Transform. The Discrete Hermite functions could be viewed as a set of orthogonal vectors that resemble a finite Fourier series. They are generated easily as eigenvectors of a symmetric tridiagonal matrix that commutes with the centered Fourier matrix. The Discrete Hermite Transform (DHmT) values are computed from a simple dot product between an individual ECG complex extracted from the European Society of Cardiology (ESC) ST-T database and the corresponding discrete Hermite function. These values are found to contain information about the ECG shape, highlighting changes between ST segment and T wave alterations which are the features of ischemic episodes. This information from the discrete Hermite transform, based on an orthonormal set of n-dimensional digital Hermite functions that serve as shape-identification functions, can be used to identify ischemic episodes from the ECG. The performance measures resulting from applying this method to detect ischemic episodes were Sensitivity 87 %, Specificity 86 %, and positive predictive accuracy 81 %. The computer time to analyze one heartbeat for ischemia with this method is 0.031 s on a standard PC.

Research on high-performance, real-time periodic signal detection method based on field-programmable gate arrays (FPGAs).

Long-Time Coherent Integration (LTCI) utilizes digital integration to combine multiple coherent cycles, thereby improving the signal-to-noise ratio (SNR). Our previous work introduced single-bit LTCI, an approach optimized for FPGA implementation, but faced challenges of output saturation at high SNR levels and inherent limitations in SNR gain (SNRG), which are insufficient for certain applications. This paper presents a threshold tracking method that improves the performance of single-bit LTCI in high-SNR scenarios. In addition, a sampling rate enhancement technique and a Kalman filtering method are introduced to further enhance the SNR of the processed signals. An FPGA-based prototype was developed to validate these methods. The results demonstrate that the threshold tracking method extends the measurable input SNR range to 30. Under the specified conditions, the sampling rate enhancement technique yields a 30% improvement in SNR over the original method, while the Kalman filter reduces noise levels to 60% of their original values.

A Hybrid Framework for Dynamic Clustering and Anomaly Detection in SAP ERP Systems

The growing complexity of enterprise data in SAP ERP systems demands advanced methods for real-time anomaly detection and dynamic clustering optimization. Traditional clustering techniques, such as Standard K-Means, need to adapt to evolving data patterns, while existing AI/ML-based methods like GMM with EM and Self-Organizing Maps (SOM) face challenges in scalability and computational efficiency. This research proposes a hybrid model that integrates GMM, SOM, and adaptive K-Means to address these limitations, offering improved detection accuracy, precision, and recall. Tested on datasets of varying sizes (small, medium, large), the hybrid model consistently outperforms baseline methods, achieving up to 94% detection accuracy and high precision (95.5%) with strong scalability and interpretability. The model leverages distributed computing on a Spark cluster for efficient processing, ensuring seamless integration with SAP ERP environments. Results demonstrate the hybrid model’s ability to detect anomalies effectively, adapt to dynamic data streams, and provide actionable insights through intuitive visualizations. This comprehensive approach makes it a robust solution for maintaining data integrity and enhancing operational decision-making in increasingly complex and data-rich enterprise environments.

AI-Powered Anomaly Detection for Real-Time Performance Monitoring in Cloud Systems

Owing to the ever growing popularity of cloud computing, there is increased need for performance monitoring coupled with real time monitoring for detection of anomalous behavior for assured availability and reliability. Nevertheless, the current practical implementation of performance monitoring in the cloud systems have certain limitations which prevent their efficient usage in the real time performance anomaly detection. The first important limitation is related to the limitations of approaches like anomalous behaviour detection in large scale and dynamically changing cloud environments, which make the task to detect such anomalies slow and highly unsuitable for large distributed systems [1], [2]. Besides, these methods apply predefined thresholds or continue data baselines, leading to higher false alarm ratio and frequent overlook of significant anomalies within altered cloud systems [3], [5]. A final problem is the rigidity of many current methods that can often fail to identify new or previously unidentified performance problems [4], [6]. Also, there is practice that requires more sophistication in the triggering of the mitigation measures to be automated to eliminate the overhead and delay in responding that cloud administrators face [7]. This paper examines these shortcomings in the existing methods for real-time anomaly detection in cloud systems, and presents the arguments for newer approaches for addressing the evolving challenges of cloud computing.

Real-time actuator fault detection and isolation for quadrotor UAV via directional residuals

Abstract Fault diagnosis is critical for the safe and reliable operation of quadrotor unmanned aerial vehicles (UAVs), so that timely remedial measures can be taken to reduce its negative impact. This paper presents a real-time actuator fault detection and isolation (FDI) method for quadrotor UAV flight process. First, a linear parameter-varying (LPV) model is established to describe quadrotor UAV dynamics, which considers measurement noise and external disturbance. Then, an FDI strategy is developed based on directional residuals, in which only one observer-based residual generator is required and Η_/L∞ indices are incorporated to balance fault sensitivity and disturbance robustness. A threshold is derived from the L∞ performance condition, and actuator fault is detected through residual evaluation. Furthermore, by introducing fault feature vectors associated with the system model, the faulty actuator is isolated promptly by analyzing directional correlations between generated residuals and the defined fault feature vectors. Finally, based on a quadrotor UAV platform, the effectiveness and superiority of the proposed FDI method are verified through online trajectory tracking processes.

BRTPillar: boosting real-time 3D object detection based point cloud and RGB image fusion in autonomous driving

PurposeIn autonomous driving, the inherent sparsity of point clouds often limits the performance of object detection, while existing multimodal architectures struggle to meet the real-time requirements for 3D object detection. Therefore, the main purpose of this paper is to significantly enhance the detection performance of objects, especially the recognition capability for small-sized objects and to address the issue of slow inference speed. This will improve the safety of autonomous driving systems and provide feasibility for devices with limited computing power to achieve autonomous driving.Design/methodology/approachBRTPillar first adopts an element-based method to fuse image and point cloud features. Secondly, a local-global feature interaction method based on an efficient additive attention mechanism was designed to extract multi-scale contextual information. Finally, an enhanced multi-scale feature fusion method was proposed by introducing adaptive spatial and channel interaction attention mechanisms, thereby improving the learning of fine-grained features.FindingsExtensive experiments were conducted on the KITTI dataset. The results showed that compared with the benchmark model, the accuracy of cars, pedestrians and cyclists on the 3D object box improved by 3.05, 9.01 and 22.65%, respectively; the accuracy in the bird’s-eye view has increased by 2.98, 10.77 and 21.14%, respectively. Meanwhile, the running speed of BRTPillar can reach 40.27 Hz, meeting the real-time detection needs of autonomous driving.Originality/valueThis paper proposes a boosting multimodal real-time 3D object detection method called BRTPillar, which achieves accurate location in many scenarios, especially for complex scenes with many small objects, while also achieving real-time inference speed.

A new quantitative real-time PCR method to measure human miRNAs using the PROMER technology

MicroRNAs (miRNAs) are small, non-coding RNA molecules that play a crucial role in regulating gene expression. Dysregulation of miRNAs is associated with various human diseases, including cancer. Accurate quantification of miRNAs in bodily fluids or tissue biopsy samples is essential for their use as biomarkers in tumor diagnosis, yet current methods remain suboptimal. In this study, we introduce a novel quantitative real-time PCR platform based on PROMER technology, which relies on RNA:DNA perfect matching in the primer-template interaction, followed by RNAse H2 cleavage to generate a 3'-hydroxyl group for new DNA synthesis.We selected 16 miRNAs implicated in lung cancer and evaluated the specificity and performance of the platform using synthesized miRNA mimics. Each miRNA could be measured with high specificity and accuracy, even at copy numbers as low as 1-10, while non-template controls exhibited minimal amplification. Additionally, key miRNAs involved in tumor progression, such as miR-210, miR-331, and miR-505, were accurately quantified using minimal amounts of plasma.In summary, we present a new quantitative real-time PCR method for miRNA detection using PROMER technology. This platform successfully measured miRNA mimics and was further applied to quantify miRNA species in plasma samples from cancer patients. Our findings support the potential clinical application of this method for early cancer diagnosis using liquid biopsy samples.

Development of TaqMan-Based Real-Time qPCR Method for Accurate Detection and Quantification of Citrus Psorosis Virus and Cytoplasmic-Type Citrus Leprosis Virus in Saplings.

In 2022, citrus fruits were the second most widely produced fruit globally, highlighting their significant role in the fruit industry. However, due to their clonal propagation, these fruits are highly susceptible to viral infections, posing challenges for growers. In response to the booming nursery market, the Korean plant quarantine station reported over 80 million sapling stocks, with 15% being discarded after rigorous inspection due to contamination or disease. As the global nursery trade continues to expand, there is an urgent need for a fast and accurate diagnostic tool to ensure the health of plant stocks. In this study, we developed a TaqMan-based real-time reverse transcription-quantitative PCR assay specifically designed to detect two critical citrus viruses: citrus psorosis virus and citrus leprosis virus C. Our assay demonstrated the capability to detect virus sequences with as few as 30 copies, maintaining high PCR efficiency with RNA extracted from both twig and leaf tissues. Additionally, we incorporated an artificial sequence into the positive controls, which effectively served as a marker for detecting potential sample contamination. This comprehensive diagnostic system promises to enhance plant quarantine measures and phytosanitation practices, providing a reliable and efficient solution to safeguard citrus crops from viral threats.

A Small Object Real-Time Detection Method for Power Line Inspection in Low-Illuminance Environments

A Small Object Real-Time Detection Method for Power Line Inspection in Low-Illuminance Environments

Vulnerability and Malware Detection

The increasingly linked digital world oftoday has made cybersecurity a top priority. The rise in cyber dangers, notably vulnerabilities and malware, poses major risks to individuals, organizations, and governments. This review article offers a thorough analysis of the approachesand instruments currently in use for malware and vulnerability detection. It examines and contrasts several detection methods, such as machine learning techniques, static and dynamic analysis, and signature-based detection. The study also examines new developments in the sector, such as the application of big data analytics and artificial intelligence (AI) to the detection of complex threats. It also covers the need for real-time detection methods, changing threat environments, and the difficulties posed by false positives.

A Bayesian Method for Real-time Unsupervised Detection of Anomalous Road Vehicle Trajectories

A Bayesian Method for Real-time Unsupervised Detection of Anomalous Road Vehicle Trajectories

Improved real-time object detection method based on YOLOv8: a refined approach

Improved real-time object detection method based on YOLOv8: a refined approach

Study on Real-Time Detection of Lightweight Tomato Plant Height Under Improved YOLOv5 and Visual Features

Tomato cultivation is relatively dense, and the main stem is easily submerged in a background environment with small color difference. The semi-enclosed planting space and fast growth cycle are both limitations that cannot be ignored in detection technology. The accuracy and real-time performance of plant height detection are of great practical significance. To this end, we are committed to improving YOLOv5 and proposing a lightweight real-time detection method for plant height by combining visual features of tomato main stems. Here, we improved the backbone, neck, head, and activation functions of YOLOv5, using CSP dark net53-s as the backbone structure and introducing a focus structure to reduce the number of GE modules. We replaced all CSP2_X structures in neck and head with GE modules, embedded interactive multi-head attention, and replaced YOLOv5’s framework function and attention activation function. We defined visual features such as the color of the main stem of tomato plants in the preprocessed image; input improved YOLOv5; and completed plant height detection through effective feature map fusion, main stem framing, and scale conversion. The experimental results show that the linear deviation between the plant height detection value and the actual value of the proposed method is always less than 3 cm, and the detection FPS can reach up to 67 frames per second, with superior timeliness, which can effectively achieve lightweight real-time detection.

X-ray fluoroscopy guided localization and steering of miniature robots using virtual reality enhancement.

In developing medical interventions using untethered milli- and microrobots, ensuring safety and effectiveness relies on robust methods for real-time robot detection, tracking, and precise localization within the body. The inherent non-transparency of human tissues significantly challenges these efforts, as traditional imaging systems like fluoroscopy often lack crucial anatomical details, potentially compromising intervention safety and efficacy. To address this technological gap, in this study, we build a virtual reality environment housing an exact digital replica (digital twin) of the operational workspace and a robot avatar. We synchronize the virtual and real workspaces and continuously send the robot position data derived from the image stream into the digital twin with short average delay time around 20-25 ms. This allows the operator to steer the robot by tracking its avatar within the digital twin with near real-time temporal resolution. We demonstrate the feasibility of this approach with millirobots steered in confined phantoms. Our concept demonstration herein can pave the way for not only improved procedural safety by complementing fluoroscopic guidance with virtual reality enhancement, but also provides a platform for incorporating various additional real-time derivative data, e.g., instantaneous robot velocity, intraoperative physiological data obtained from the patient, e.g., blood flow rate, and pre-operative physical simulation models, e.g., periodic body motions, to further refine robot control capacity.

An efficient re-parameterization feature pyramid network on YOLOv8 to the detection of steel surface defect

In the field of steel production, the detection of steel surface defects is one of the most important guarantees for the quality of steel production. In the process of defect detection, there are problems regarding the noise of the acquisition background, the scale of defects, and the detection speed. At present, in the face of complex steel surface defects, realizing efficient real-time steel surface defect detection has become a difficult problem. In this paper, we propose a lightweight and efficient real-time defect detection method, LDE-YOLO, based on YOLOv8. First, we propose a lightweight multi-scale feature extraction module, LighterMSMC, which not only achieves a lightweight backbone network, but also effectively guarantees the long range dependence of the features, so as to realize multi-scale feature extraction more efficiently. Secondly, we propose lightweight re-parameterized feature pyramid, DE-FPN, in which the sparse patterns of the overall features and the detailed features of the local features are efficiently captured by the DE-Block, and then efficiently fused by the PAN feature fusion structure. Finally, we propose Efficient Head, which lightens the model by group convolution while its improves the diagonal correlation of the feature maps on some specific datasets, thus enhancing the detection performance. Our proposed LDE-YOLO obtains 80.8 mAP and 75.5 FPS on NEU-DET , 80.5 mAP and 75.5 FPS on GC10-DET. It obtains 2.5 mAP and 4.7 mAP enhancement compared to the baseline model, and the detection speed is also improved by 10.4 FPS, while in terms of the number of floating point operations and parameters of the model reduced by 60.2% and 49.1%, which is sufficient to illustrate its lightweight effectiveness and realize an efficient real-time steel surface defect detection model.

Real-time object detection method with single-domain generalization based on YOLOv8

Real-time object detection method with single-domain generalization based on YOLOv8