End-to-end Temporal Action Detection(TAD)has achieved remarkable progress in recent years,driven by innovations in model architectures and the emergence of Video Foundation Models(VFMs).However,existing TAD methods th...End-to-end Temporal Action Detection(TAD)has achieved remarkable progress in recent years,driven by innovations in model architectures and the emergence of Video Foundation Models(VFMs).However,existing TAD methods that perform full fine-tuning of pretrained video models often incur substantial computational costs,which become particularly pronounced when processing long video sequences.Moreover,the need for precise temporal boundary annotations makes data labeling extremely expensive.In low-resource settings where annotated samples are scarce,direct fine-tuning tends to cause overfitting.To address these challenges,we introduce Dynamic LowRank Adapter(DyLoRA),a lightweight fine-tuning framework tailored specifically for the TAD task.Built upon the Low-Rank Adaptation(LoRA)architecture,DyLoRA adapts only the key layers of the pretrained model via low-rank decomposition,reducing the number of trainable parameters to less than 5%of full fine-tuning methods.This significantly lowers memory consumption and mitigates overfitting in low-resource settings.Notably,DyLoRA enhances the temporal modeling capability of pretrained models by optimizing temporal dimension weights,thereby alleviating the representation misalignment of temporal features.Experimental results demonstrate that DyLoRA-TAD achieves impressive performance,with 73.9%mAP on THUMOS14,39.52%on ActivityNet-1.3,and 28.2%on Charades,substantially surpassing the best traditional feature-based methods.展开更多
The growing incidence of cyberattacks necessitates a robust and effective Intrusion Detection Systems(IDS)for enhanced network security.While conventional IDSs can be unsuitable for detecting different and emerging at...The growing incidence of cyberattacks necessitates a robust and effective Intrusion Detection Systems(IDS)for enhanced network security.While conventional IDSs can be unsuitable for detecting different and emerging attacks,there is a demand for better techniques to improve detection reliability.This study introduces a new method,the Deep Adaptive Multi-Layer Attention Network(DAMLAN),to boost the result of intrusion detection on network data.Due to its multi-scale attention mechanisms and graph features,DAMLAN aims to address both known and unknown intrusions.The real-world NSL-KDD dataset,a popular choice among IDS researchers,is used to assess the proposed model.There are 67,343 normal samples and 58,630 intrusion attacks in the training set,12,833 normal samples,and 9711 intrusion attacks in the test set.Thus,the proposed DAMLAN method is more effective than the standard models due to the consideration of patterns by the attention layers.The experimental performance of the proposed model demonstrates that it achieves 99.26%training accuracy and 90.68%testing accuracy,with precision reaching 98.54%on the training set and 96.64%on the testing set.The recall and F1 scores again support the model with training set values of 99.90%and 99.21%and testing set values of 86.65%and 91.37%.These results provide a strong basis for the claims made regarding the model’s potential to identify intrusion attacks and affirm its relatively strong overall performance,irrespective of type.Future work would employ more attempts to extend the scalability and applicability of DAMLAN for real-time use in intrusion detection systems.展开更多
In the context of target detection under infrared conditions for drones,the common issues of high missed detection rates,low signal-to-noise ratio,and blurred edge features for small targets are prevalent.To address t...In the context of target detection under infrared conditions for drones,the common issues of high missed detection rates,low signal-to-noise ratio,and blurred edge features for small targets are prevalent.To address these challenges,this paper proposes an improved detection algorithm based on YOLOv11n.First,a Dynamic Multi-Scale Feature Fusion and Adaptive Weighting approach is employed to design an Adaptive Focused Diffusion Pyramid Network(AFDPN),which enhances the feature expression and transmission capability of shallow small targets,thereby reducing the loss of detailed information.Then,combined with an Edge Enhancement(EE)module,the model improves the extraction of infrared small target edge features through low-frequency suppression and high-frequency enhancement strategies.Experimental results on the publicly available HIT-UAV dataset show that the improved model achieves a 3.8%increase in average detection accuracy and a 3.0%improvement in recall rate compared to YOLOv11n,with a computational cost of only 9.1 GFLOPS.In comparison experiments,the detection accuracy and model size balance achieved the optimal solution,meeting the lightweight deployment requirements for drone-based systems.This method provides a high-precision,lightweight solution for small target detection in drone-based infrared imagery.展开更多
Global Navigation Satellite Systems(GNSSs)face significant security threats from spoofing attacks.Typical anti-spoofing methods rely on estimating the delays between spoofing and authentic signals using multicorrelato...Global Navigation Satellite Systems(GNSSs)face significant security threats from spoofing attacks.Typical anti-spoofing methods rely on estimating the delays between spoofing and authentic signals using multicorrelator outputs.However,the accuracy of the delay estimation is limited by the spacing of the correlators.To address this,an innovative anti-spoofing method is introduced,which incorporates distinct coarse and refined stages for more accurate spoofing estimation.By leveraging the coarse delay estimates obtained through maximum likelihood estimation,the proposed method establishes the Windowed Sum of the Relative Delay(WSRD)statistics to detect the presence of spoofing signals.The iterative strategy is then employed to enhance the precision of the delay estimation.To further adapt to variations in the observation noise caused by spoofing intrusions and restore precise position,velocity,and timing solutions,an adaptive extended Kalman filter is proposed.This comprehensive framework offers detection,mitigation,and recovery against spoofing attacks.Experimental validation using datasets from the Texas Spoofing Test Battery(TEXBAT)demonstrates the effectiveness of the proposed anti-spoofing method.With 41 correlators,the method achieves a detection rate exceeding 90%at a false alarm rate of 10-5,with position or time errors below 15 m.Notably,this refined anti-spoofing approach shows robust detection and mitigation capabilities,requiring only a single antenna without the need for additional external sensors.These advancements can significantly contribute to the development of GNSS anti-spoofing measures.展开更多
Adaptive robust secure framework plays a vital role in implementing intelligent automation and decentralized decision making of Industry 5.0.Latency,privacy risks and the complexity of industrial networks have been pr...Adaptive robust secure framework plays a vital role in implementing intelligent automation and decentralized decision making of Industry 5.0.Latency,privacy risks and the complexity of industrial networks have been preventing attempts at traditional cloud-based learning systems.We demonstrate that,to overcome these challenges,for instance,the EdgeGuard-IoT framework,a 6G edge intelligence framework enhancing cybersecurity and operational resilience of the smart grid,is needed on the edge to integrate Secure Federated Learning(SFL)and Adaptive Anomaly Detection(AAD).With ultra-reliable low latency communication(URLLC)of 6G,artificial intelligence-based network orchestration,and massive machine type communication(mMTC),EdgeGuard-IoT brings real-time,distributed intelligence on the edge,and mitigates risks in data transmission and enhances privacy.EdgeGuard-IoT,with a hierarchical federated learning framework,helps edge devices to collaboratively train models without revealing the sensitive grid data,which is crucial in the smart grid where real-time power anomaly detection and the decentralization of the energy management are a big deal.The hybrid AI models driven adaptive anomaly detection mechanism immediately raises the thumb if the grid stability and strength are negatively affected due to cyber threats,faults,and energy distribution,thereby keeping the grid stable with resilience.The proposed framework also adopts various security means within the blockchain and zero-trust authentication techniques to reduce the adversarial attack risks and model poisoning during federated learning.EdgeGuard-IoT shows superior detection accuracy,response time,and scalability performance at a much reduced communication overhead via extensive simulations and deployment in real-world case studies in smart grids.This research pioneers a 6G-driven federated intelligence model designed for secure,self-optimizing,and resilient Industry 5.0 ecosystems,paving the way for next-generation autonomous smart grids and industrial cyber-physical systems.展开更多
Cardiovascular diseases(CVDs)continue to present a leading cause ofmortalityworldwide,emphasizing the importance of early and accurate prediction.Electrocardiogram(ECG)signals,central to cardiac monitoring,have increa...Cardiovascular diseases(CVDs)continue to present a leading cause ofmortalityworldwide,emphasizing the importance of early and accurate prediction.Electrocardiogram(ECG)signals,central to cardiac monitoring,have increasingly been integratedwithDeep Learning(DL)for real-time prediction of CVDs.However,DL models are prone to performance degradation due to concept drift and to catastrophic forgetting.To address this issue,we propose a realtime CVDs prediction approach,referred to as ADWIN-GFR that combines Convolutional Neural Network(CNN)layers,for spatial feature extraction,with Gated Recurrent Units(GRU),for temporal modeling,alongside adaptive drift detection and mitigation mechanisms.The proposed approach integratesAdaptiveWindowing(ADWIN)for realtime concept drift detection,a fine-tuning strategy based on Generative Features Replay(GFR)to preserve previously acquired knowledge,and a dynamic replay buffer ensuring variance,diversity,and data distribution coverage.Extensive experiments conducted on the MIT-BIH arrhythmia dataset demonstrate that ADWIN-GFR outperforms standard fine-tuning techniques,achieving an average post-drift accuracy of 95.4%,amacro F1-score of 93.9%,and a remarkably low forgetting score of 0.9%.It also exhibits an average drift detection delay of 12 steps and achieves an adaptation gain of 17.2%.These findings underscore the potential of ADWIN-GFR for deployment in real-world cardiac monitoring systems,including wearable ECG devices and hospital-based patient monitoring platforms.展开更多
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 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.展开更多
Generative Adversarial Networks(GANs)have become valuable tools in medical imaging,enabling realistic image synthesis for enhancement,augmentation,and restoration.However,their integration into clinical workflows rais...Generative Adversarial Networks(GANs)have become valuable tools in medical imaging,enabling realistic image synthesis for enhancement,augmentation,and restoration.However,their integration into clinical workflows raises concerns,particularly the risk of subtle distortions or hallucinations that may undermine diagnostic accuracy and weaken trust in AI-assisted decision-making.To address this challenge,we propose a hybrid deep learning framework designed to detect GAN-induced artifacts in medical images,thereby reinforcing the reliability of AI-driven diagnostics.The framework integrates low-level statistical descriptors,including high-frequency residuals and Gray-Level Co-occurrence Matrix(GLCM)texture features,with high-level semantic representations extracted from a pre-trained ResNet18.This dual-stream approach enables detection of both pixel-level anomalies and structural inconsistencies introduced by GAN-based manipulation.We validated the framework on a curated dataset of 10,000 medical images,evenly split between authentic and GAN-generated samples across four modalities:MRI,CT,X-ray,and fundus photography.To improve generalizability to real-world clinical settings,we incorporated domain adaptation strategies such as adversarial training and style transfer,reducing domain shift by 15%.Experimental results demonstrate robust performance,achieving 92.6%accuracy and an F1-score of 0.91 on synthetic test data,and maintaining strong performance on real-world GAN-modified images with 87.3%accuracy and an F1-score of 0.85.Additionally,the model attained an AUC of 0.96 and an average precision of 0.92,outperforming conventional GAN detection pipelines and baseline Convolutional Neural Network(CNN)architectures.These findings establish the proposed framework as an effective and reliable solution for detecting GAN-induced hallucinations in medical imaging,representing an important step toward building trustworthy and clinically deployable AI systems.展开更多
The current morphological wavelet technologies utilize a fixed filter or a linear decomposition algorithm, which cannot cope with the sudden changes, such as impulses or edges in a signal effectively. This paper pre- ...The current morphological wavelet technologies utilize a fixed filter or a linear decomposition algorithm, which cannot cope with the sudden changes, such as impulses or edges in a signal effectively. This paper pre- sents a novel signal processing scheme, adaptive morpho- logical update lifting wavelet （AMULW）, for rolling element bearing fault detection. In contrast with the widely used morphological wavelet, the filters in AMULW are no longer fixed. Instead, the AMULW adaptively uses a morphological dilation-erosion filter or an average filter as the update lifting filter to modify the approximation signal. Moreover, the nonlinear morphological filter is utilized to substitute the traditional linear filter in AMULW. The effectiveness of the proposed AMULW is evaluated using a simulated vibration signal and experimental vibration sig- nals collected from a bearing test rig. Results show that the proposed method has a superior performance in extracting fault features of defective roiling element bearings.展开更多
A method of woven fabric defect detection using the wavelet transform adaptive to the fabric has been developed. With reference to the orthogonality constrains of Daubechies wavelet, by taking the mmimization of the e...A method of woven fabric defect detection using the wavelet transform adaptive to the fabric has been developed. With reference to the orthogonality constrains of Daubechies wavelet, by taking the mmimization of the energy or the gray level of the pixels in the output sub-images as the additional conditions and using the random algorithm method, two sets of wavelet filters adapted to the fabric texture were formed. The original images of normal fabric texture and the fabric texture with defects were decomposed into horizontal and vertical sub- images by using these filters and the feature indices of these sub-images were also extracted. By comparing the feature indices of the normal texture with that of the defective texture, the fabric defects can be successfully detected and located.展开更多
As a primary defense technique, intrusion detection becomes more and more significant since the security of the networks is one of the most critical issues in the world. We present an adaptive collaboration intrusion ...As a primary defense technique, intrusion detection becomes more and more significant since the security of the networks is one of the most critical issues in the world. We present an adaptive collaboration intrusion detection method to improve the safety of a network. A self-adaptive and collaborative intrusion detection model is built by applying the Environmentsclasses, agents, roles, groups, and objects(E-CARGO) model. The objects, roles, agents, and groups are designed by using decision trees(DTs) and support vector machines(SVMs), and adaptive scheduling mechanisms are set up. The KDD CUP 1999 data set is used to verify the effectiveness of the method. The experimental results demonstrate the feasibility and efficiency of the proposed collaborative and adaptive intrusion detection method. Also, the proposed method is shown to be more predominant than the methods that use a set of single type support vector machine(SVM) in terms of detection precision rate and recall rate.展开更多
Since the atmospheric correction is a necessary preprocessing step of remote sensing image before detecting green tide, the introduced error directly affects the detection precision. Therefore, the detection method of...Since the atmospheric correction is a necessary preprocessing step of remote sensing image before detecting green tide, the introduced error directly affects the detection precision. Therefore, the detection method of green tide is presented from Landsat TM/ETM plus image which needs not the atmospheric correction. In order to achieve an automatic detection of green tide, a linear relationship（y =0.723 x＋0.504） between detection threshold y and subtraction x（x=λnir–λred） is found from the comparing Landsat TM/ETM plus image with the field surveys.Using this relationship, green tide patches can be detected automatically from Landsat TM/ETM plus image.Considering there is brightness difference between different regions in an image, the image will be divided into a plurality of windows（sub-images） with a same size firstly, and then each window will be detected using an adaptive detection threshold determined according to the discovered linear relationship. It is found that big errors will appear in some windows, such as those covered by clouds seriously. To solve this problem, the moving step k of windows is proposed to be less than the window width n. Using this mechanism, most pixels will be detected[n/k]×[n/k] times except the boundary pixels, then every pixel will be assigned the final class（green tide or sea water） according to majority rule voting strategy. It can be seen from the experiments, the proposed detection method using multi-windows and their adaptive thresholds can detect green tide from Landsat TM/ETM plus image automatically. Meanwhile, it avoids the reliance on the accurate atmospheric correction.展开更多
With the gradual penetration of the internet of things(IoT)into all areas of life,the scale of IoT devices shows an explosive growth trend.The era of internet of everything is coming,and the important position of IoT ...With the gradual penetration of the internet of things(IoT)into all areas of life,the scale of IoT devices shows an explosive growth trend.The era of internet of everything is coming,and the important position of IoT security is becoming increasingly prominent.Due to the large number types of IoT devices,there may be different security vulnerabilities,and unknown attack forms and virus samples are appear.In other words,large number of IoT devices,large data volumes,and various attack forms pose a big challenge of malicious traffic identification.To solve these problems,this paper proposes a concept drift detection and adaptation(CDDA)method for IoT security framework.The AI model performance is evaluated by verifying the effectiveness of IoT traffic for data drift detection,so as to select the best AI model.The experimental test are given to confirm that the feasibility of the framework and the adaptive method in practice,and the effect on the performance of IoT traffic identification is also verified.展开更多
Visual background extraction algorithm(ViBe)uses the first frame image to initialize the background model,which can easily introduce the“ghost”.Because ViBe uses the fixed segmentation threshold to achieve the foreg...Visual background extraction algorithm(ViBe)uses the first frame image to initialize the background model,which can easily introduce the“ghost”.Because ViBe uses the fixed segmentation threshold to achieve the foreground and background segmentation,the detection results in many false detections for the highly dynamic background.To solve these problems,an improved ghost suppression and adaptive Visual Background Extraction algorithm is proposed in this paper.Firstly,with the pixel’s temporal and spatial information,the historical pixels of a certain combination are used to initialize the background model in the odd frames of the video sequence.Secondly,the background sample set combined with the neighborhood pixels are used to determine a complex degree of the background,to acquire the adaptive segmentation threshold.Thirdly,the update rate is adjusted based on the complexity of the background.Finally,the detected result goes through a post-processing to achieve better detection results.The experimental results show that the improved algorithm will not only quickly suppress the“ghost”,but also have a better detection in a complex dynamic background.展开更多
In chemical process, a large number of measured and manipulated variables are highly correlated. Principal component analysis(PCA) is widely applied as a dimension reduction technique for capturing strong correlation ...In chemical process, a large number of measured and manipulated variables are highly correlated. Principal component analysis(PCA) is widely applied as a dimension reduction technique for capturing strong correlation underlying in the process measurements. However, it is difficult for PCA based fault detection results to be interpreted physically and to provide support for isolation. Some approaches incorporating process knowledge are developed, but the information is always shortage and deficient in practice. Therefore, this work proposes an adaptive partitioning PCA algorithm entirely based on operation data. The process feature space is partitioned into several sub-feature spaces. Constructed sub-block models can not only reflect the local behavior of process change, namely to grasp the intrinsic local information underlying the process changes, but also improve the fault detection and isolation through the combination of local fault detection results and reduction of smearing effect.The method is demonstrated in TE process, and the results show that the new method is much better in fault detection and isolation compared to conventional PCA method.展开更多
Cone photoreceptor cell identication is important for the early diagnosis of retinopathy.In this study,an object detection algorithm is used for cone cell identication in confocal adaptive optics scanning laser ophtha...Cone photoreceptor cell identication is important for the early diagnosis of retinopathy.In this study,an object detection algorithm is used for cone cell identication in confocal adaptive optics scanning laser ophthalmoscope(AOSLO)images.An effectiveness evaluation of identication using the proposed method reveals precision,recall,and F_(1)-score of 95.8%,96.5%,and 96.1%,respectively,considering manual identication as the ground truth.Various object detection and identication results from images with different cone photoreceptor cell distributions further demonstrate the performance of the proposed method.Overall,the proposed method can accurately identify cone photoreceptor cells on confocal adaptive optics scanning laser ophthalmoscope images,being comparable to manual identication.展开更多
Low Resolution Thermal Array Sensors are widely used in several applications in indoor environments. In particular, one of these cheap, small and unobtrusive sensors provides a low-resolution thermal image of the envi...Low Resolution Thermal Array Sensors are widely used in several applications in indoor environments. In particular, one of these cheap, small and unobtrusive sensors provides a low-resolution thermal image of the environment and, unlike cameras;it is capable to detect human heat emission even in dark rooms. The obtained thermal data can be used to monitor older seniors while they are performing daily activities at home, to detect critical situations such as falls. Most of the studies in activity recognition using Thermal Array Sensors require human detection techniques to recognize humans passing in the sensor field of view. This paper aims to improve the accuracy of the algorithms used so far by considering the temperature environment variation. This method leverages an adaptive background estimation and a noise removal technique based on Kalman Filter. In order to properly validate the system, a novel installation of a single sensor has been implemented in a smart environment: the obtained results show an improvement in human detection accuracy with respect to the state of the art, especially in case of disturbed environments.展开更多
This study aimed to address the challenge of accurately and reliably detecting tomatoes in dense planting environments,a critical prerequisite for the automation implementation of robotic harvesting.However,the heavy ...This study aimed to address the challenge of accurately and reliably detecting tomatoes in dense planting environments,a critical prerequisite for the automation implementation of robotic harvesting.However,the heavy reliance on extensive manually annotated datasets for training deep learning models still poses significant limitations to their application in real-world agricultural production environments.To overcome these limitations,we employed domain adaptive learning approach combined with the YOLOv5 model to develop a novel tomato detection model called as TDA-YOLO(tomato detection domain adaptation).We designated the normal illumination scenes in dense planting environments as the source domain and utilized various other illumination scenes as the target domain.To construct bridge mechanism between source and target domains,neural preset for color style transfer is introduced to generate a pseudo-dataset,which served to deal with domain discrepancy.Furthermore,this study combines the semi-supervised learning method to enable the model to extract domain-invariant features more fully,and uses knowledge distillation to improve the model's ability to adapt to the target domain.Additionally,for purpose of promoting inference speed and low computational demand,the lightweight FasterNet network was integrated into the YOLOv5's C3 module,creating a modified C3_Faster module.The experimental results demonstrated that the proposed TDA-YOLO model significantly outperformed original YOLOv5s model,achieving a mAP(mean average precision)of 96.80%for tomato detection across diverse scenarios in dense planting environments,increasing by 7.19 percentage points;Compared with the latest YOLOv8 and YOLOv9,it is also 2.17 and 1.19 percentage points higher,respectively.The model's average detection time per image was an impressive 15 milliseconds,with a FLOPs(floating point operations per second)count of 13.8 G.After acceleration processing,the detection accuracy of the TDA-YOLO model on the Jetson Xavier NX development board is 90.95%,the mAP value is 91.35%,and the detection time of each image is 21 ms,which can still meet the requirements of real-time detection of tomatoes in dense planting environment.The experimental results show that the proposed TDA-YOLO model can accurately and quickly detect tomatoes in dense planting environment,and at the same time avoid the use of a large number of annotated data,which provides technical support for the development of automatic harvesting systems for tomatoes and other fruits.展开更多
An adaptive endpoint detection algorithm based on band energy and adaptive smoothing algorithm is described. This algorithm utilizes the capability of adaptive smoothing algorithm that intensifies the discontinuity be...An adaptive endpoint detection algorithm based on band energy and adaptive smoothing algorithm is described. This algorithm utilizes the capability of adaptive smoothing algorithm that intensifies the discontinuity between local areas. The band energy features are selected because of their usefulness in detecting high energy regions (in the incoming signal) and making the distinction between speech and noise. Heuristic 'edge-focusing' is used to endpoint detection to save the time in iteration.展开更多
The detection performance and the constant false alarm rate behavior of the conventional adaptive detectors are severely degraded in heterogeneous clutter. This paper designs and analyses a knowledge-based （KB） adap...The detection performance and the constant false alarm rate behavior of the conventional adaptive detectors are severely degraded in heterogeneous clutter. This paper designs and analyses a knowledge-based （KB） adaptive polarimetric detector in het-erogeneous clutter. The proposed detection scheme is composed of a data selector using polarization knowledge and an adaptive polarization detector using training data. A polarization data selector based on the maximum likelihood estimation is proposed to remove outliers from the heterogeneous training data. This selector can remove outliers effectively, thus the training data is purified for estimating the clutter covariance matrix. Consequently, the performance of the adaptive detector is improved. We assess the performance of the KB adaptive polarimetric detector and the adaptive polarimetric detector without a data selector using simulated data and IPIX radar data. The results show that the KB adaptive polarization detector outperforms its non-KB counterparts.展开更多
基金supported by the National Natural Science Foundation of China(Grant No.62266054)the Major Science and Technology Project of Yunnan Province(Grant No.202402AD080002)the Scientific Research Fund of the Yunnan Provincial Department of Education(Grant No.2025Y0302).
文摘End-to-end Temporal Action Detection(TAD)has achieved remarkable progress in recent years,driven by innovations in model architectures and the emergence of Video Foundation Models(VFMs).However,existing TAD methods that perform full fine-tuning of pretrained video models often incur substantial computational costs,which become particularly pronounced when processing long video sequences.Moreover,the need for precise temporal boundary annotations makes data labeling extremely expensive.In low-resource settings where annotated samples are scarce,direct fine-tuning tends to cause overfitting.To address these challenges,we introduce Dynamic LowRank Adapter(DyLoRA),a lightweight fine-tuning framework tailored specifically for the TAD task.Built upon the Low-Rank Adaptation(LoRA)architecture,DyLoRA adapts only the key layers of the pretrained model via low-rank decomposition,reducing the number of trainable parameters to less than 5%of full fine-tuning methods.This significantly lowers memory consumption and mitigates overfitting in low-resource settings.Notably,DyLoRA enhances the temporal modeling capability of pretrained models by optimizing temporal dimension weights,thereby alleviating the representation misalignment of temporal features.Experimental results demonstrate that DyLoRA-TAD achieves impressive performance,with 73.9%mAP on THUMOS14,39.52%on ActivityNet-1.3,and 28.2%on Charades,substantially surpassing the best traditional feature-based methods.
基金Nourah bint Abdulrahman University for funding this project through the Researchers Supporting Project(PNURSP2025R319)Riyadh,Saudi Arabia and Prince Sultan University for covering the article processing charges(APC)associated with this publication.Special acknowledgement to Automated Systems&Soft Computing Lab(ASSCL),Prince Sultan University,Riyadh,Saudi Arabia.
文摘The growing incidence of cyberattacks necessitates a robust and effective Intrusion Detection Systems(IDS)for enhanced network security.While conventional IDSs can be unsuitable for detecting different and emerging attacks,there is a demand for better techniques to improve detection reliability.This study introduces a new method,the Deep Adaptive Multi-Layer Attention Network(DAMLAN),to boost the result of intrusion detection on network data.Due to its multi-scale attention mechanisms and graph features,DAMLAN aims to address both known and unknown intrusions.The real-world NSL-KDD dataset,a popular choice among IDS researchers,is used to assess the proposed model.There are 67,343 normal samples and 58,630 intrusion attacks in the training set,12,833 normal samples,and 9711 intrusion attacks in the test set.Thus,the proposed DAMLAN method is more effective than the standard models due to the consideration of patterns by the attention layers.The experimental performance of the proposed model demonstrates that it achieves 99.26%training accuracy and 90.68%testing accuracy,with precision reaching 98.54%on the training set and 96.64%on the testing set.The recall and F1 scores again support the model with training set values of 99.90%and 99.21%and testing set values of 86.65%and 91.37%.These results provide a strong basis for the claims made regarding the model’s potential to identify intrusion attacks and affirm its relatively strong overall performance,irrespective of type.Future work would employ more attempts to extend the scalability and applicability of DAMLAN for real-time use in intrusion detection systems.
文摘In the context of target detection under infrared conditions for drones,the common issues of high missed detection rates,low signal-to-noise ratio,and blurred edge features for small targets are prevalent.To address these challenges,this paper proposes an improved detection algorithm based on YOLOv11n.First,a Dynamic Multi-Scale Feature Fusion and Adaptive Weighting approach is employed to design an Adaptive Focused Diffusion Pyramid Network(AFDPN),which enhances the feature expression and transmission capability of shallow small targets,thereby reducing the loss of detailed information.Then,combined with an Edge Enhancement(EE)module,the model improves the extraction of infrared small target edge features through low-frequency suppression and high-frequency enhancement strategies.Experimental results on the publicly available HIT-UAV dataset show that the improved model achieves a 3.8%increase in average detection accuracy and a 3.0%improvement in recall rate compared to YOLOv11n,with a computational cost of only 9.1 GFLOPS.In comparison experiments,the detection accuracy and model size balance achieved the optimal solution,meeting the lightweight deployment requirements for drone-based systems.This method provides a high-precision,lightweight solution for small target detection in drone-based infrared imagery.
基金co-supported by the Tianjin Research innovation Project for Postgraduate Students,China(No.2022BKYZ039)the China Postdoctoral Science Foundation(No.2023M731788)the National Natural Science Foundation of China(No.62303246)。
文摘Global Navigation Satellite Systems(GNSSs)face significant security threats from spoofing attacks.Typical anti-spoofing methods rely on estimating the delays between spoofing and authentic signals using multicorrelator outputs.However,the accuracy of the delay estimation is limited by the spacing of the correlators.To address this,an innovative anti-spoofing method is introduced,which incorporates distinct coarse and refined stages for more accurate spoofing estimation.By leveraging the coarse delay estimates obtained through maximum likelihood estimation,the proposed method establishes the Windowed Sum of the Relative Delay(WSRD)statistics to detect the presence of spoofing signals.The iterative strategy is then employed to enhance the precision of the delay estimation.To further adapt to variations in the observation noise caused by spoofing intrusions and restore precise position,velocity,and timing solutions,an adaptive extended Kalman filter is proposed.This comprehensive framework offers detection,mitigation,and recovery against spoofing attacks.Experimental validation using datasets from the Texas Spoofing Test Battery(TEXBAT)demonstrates the effectiveness of the proposed anti-spoofing method.With 41 correlators,the method achieves a detection rate exceeding 90%at a false alarm rate of 10-5,with position or time errors below 15 m.Notably,this refined anti-spoofing approach shows robust detection and mitigation capabilities,requiring only a single antenna without the need for additional external sensors.These advancements can significantly contribute to the development of GNSS anti-spoofing measures.
基金supported by Department of Information Technology,University of Tabuk,Tabuk,71491,Saudi Arabia.
文摘Adaptive robust secure framework plays a vital role in implementing intelligent automation and decentralized decision making of Industry 5.0.Latency,privacy risks and the complexity of industrial networks have been preventing attempts at traditional cloud-based learning systems.We demonstrate that,to overcome these challenges,for instance,the EdgeGuard-IoT framework,a 6G edge intelligence framework enhancing cybersecurity and operational resilience of the smart grid,is needed on the edge to integrate Secure Federated Learning(SFL)and Adaptive Anomaly Detection(AAD).With ultra-reliable low latency communication(URLLC)of 6G,artificial intelligence-based network orchestration,and massive machine type communication(mMTC),EdgeGuard-IoT brings real-time,distributed intelligence on the edge,and mitigates risks in data transmission and enhances privacy.EdgeGuard-IoT,with a hierarchical federated learning framework,helps edge devices to collaboratively train models without revealing the sensitive grid data,which is crucial in the smart grid where real-time power anomaly detection and the decentralization of the energy management are a big deal.The hybrid AI models driven adaptive anomaly detection mechanism immediately raises the thumb if the grid stability and strength are negatively affected due to cyber threats,faults,and energy distribution,thereby keeping the grid stable with resilience.The proposed framework also adopts various security means within the blockchain and zero-trust authentication techniques to reduce the adversarial attack risks and model poisoning during federated learning.EdgeGuard-IoT shows superior detection accuracy,response time,and scalability performance at a much reduced communication overhead via extensive simulations and deployment in real-world case studies in smart grids.This research pioneers a 6G-driven federated intelligence model designed for secure,self-optimizing,and resilient Industry 5.0 ecosystems,paving the way for next-generation autonomous smart grids and industrial cyber-physical systems.
基金supported by Princess Nourah bint Abdulrahman University Researchers Supporting Project number(PNURSP2025R196)Princess Nourah bint Abdulrahman University,Riyadh,Saudi Arabia.
文摘Cardiovascular diseases(CVDs)continue to present a leading cause ofmortalityworldwide,emphasizing the importance of early and accurate prediction.Electrocardiogram(ECG)signals,central to cardiac monitoring,have increasingly been integratedwithDeep Learning(DL)for real-time prediction of CVDs.However,DL models are prone to performance degradation due to concept drift and to catastrophic forgetting.To address this issue,we propose a realtime CVDs prediction approach,referred to as ADWIN-GFR that combines Convolutional Neural Network(CNN)layers,for spatial feature extraction,with Gated Recurrent Units(GRU),for temporal modeling,alongside adaptive drift detection and mitigation mechanisms.The proposed approach integratesAdaptiveWindowing(ADWIN)for realtime concept drift detection,a fine-tuning strategy based on Generative Features Replay(GFR)to preserve previously acquired knowledge,and a dynamic replay buffer ensuring variance,diversity,and data distribution coverage.Extensive experiments conducted on the MIT-BIH arrhythmia dataset demonstrate that ADWIN-GFR outperforms standard fine-tuning techniques,achieving an average post-drift accuracy of 95.4%,amacro F1-score of 93.9%,and a remarkably low forgetting score of 0.9%.It also exhibits an average drift detection delay of 12 steps and achieves an adaptation gain of 17.2%.These findings underscore the potential of ADWIN-GFR for deployment in real-world cardiac monitoring systems,including wearable ECG devices and hospital-based patient monitoring platforms.
基金The National Natural Science Foundation of China (No.61172135,61101198)the Aeronautical Foundation of China (No.20115152026)
文摘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.
基金supported and funded by the Deanship of Scientific Research at Imam Mohammad Ibn Saud Islamic University(IMSIU)(grant number IMSIU-DDRSP2601).
文摘Generative Adversarial Networks(GANs)have become valuable tools in medical imaging,enabling realistic image synthesis for enhancement,augmentation,and restoration.However,their integration into clinical workflows raises concerns,particularly the risk of subtle distortions or hallucinations that may undermine diagnostic accuracy and weaken trust in AI-assisted decision-making.To address this challenge,we propose a hybrid deep learning framework designed to detect GAN-induced artifacts in medical images,thereby reinforcing the reliability of AI-driven diagnostics.The framework integrates low-level statistical descriptors,including high-frequency residuals and Gray-Level Co-occurrence Matrix(GLCM)texture features,with high-level semantic representations extracted from a pre-trained ResNet18.This dual-stream approach enables detection of both pixel-level anomalies and structural inconsistencies introduced by GAN-based manipulation.We validated the framework on a curated dataset of 10,000 medical images,evenly split between authentic and GAN-generated samples across four modalities:MRI,CT,X-ray,and fundus photography.To improve generalizability to real-world clinical settings,we incorporated domain adaptation strategies such as adversarial training and style transfer,reducing domain shift by 15%.Experimental results demonstrate robust performance,achieving 92.6%accuracy and an F1-score of 0.91 on synthetic test data,and maintaining strong performance on real-world GAN-modified images with 87.3%accuracy and an F1-score of 0.85.Additionally,the model attained an AUC of 0.96 and an average precision of 0.92,outperforming conventional GAN detection pipelines and baseline Convolutional Neural Network(CNN)architectures.These findings establish the proposed framework as an effective and reliable solution for detecting GAN-induced hallucinations in medical imaging,representing an important step toward building trustworthy and clinically deployable AI systems.
基金Supported by National Natural Science Foundation of China(51705431,51375078)Natural Sciences and Engineering Research Council of Canada(RGPIN-2015-04897)
文摘The current morphological wavelet technologies utilize a fixed filter or a linear decomposition algorithm, which cannot cope with the sudden changes, such as impulses or edges in a signal effectively. This paper pre- sents a novel signal processing scheme, adaptive morpho- logical update lifting wavelet （AMULW）, for rolling element bearing fault detection. In contrast with the widely used morphological wavelet, the filters in AMULW are no longer fixed. Instead, the AMULW adaptively uses a morphological dilation-erosion filter or an average filter as the update lifting filter to modify the approximation signal. Moreover, the nonlinear morphological filter is utilized to substitute the traditional linear filter in AMULW. The effectiveness of the proposed AMULW is evaluated using a simulated vibration signal and experimental vibration sig- nals collected from a bearing test rig. Results show that the proposed method has a superior performance in extracting fault features of defective roiling element bearings.
基金This research was supported by the Research Fund for the Doctoral Program of Higher Education, No.99025508
文摘A method of woven fabric defect detection using the wavelet transform adaptive to the fabric has been developed. With reference to the orthogonality constrains of Daubechies wavelet, by taking the mmimization of the energy or the gray level of the pixels in the output sub-images as the additional conditions and using the random algorithm method, two sets of wavelet filters adapted to the fabric texture were formed. The original images of normal fabric texture and the fabric texture with defects were decomposed into horizontal and vertical sub- images by using these filters and the feature indices of these sub-images were also extracted. By comparing the feature indices of the normal texture with that of the defective texture, the fabric defects can be successfully detected and located.
基金supported in part by the National Natural Science Foundation of China(61772141,61673123)Guangdong Provincial Science&Technology Project(2015B090901016,2016B010108007)+1 种基金Guangdong Education Department Project(Guangdong Higher Education letter 2015[133])the Guangzhou Science&Technology Project(201508010067,201604020145201604046017,and 2016201604030034)
文摘As a primary defense technique, intrusion detection becomes more and more significant since the security of the networks is one of the most critical issues in the world. We present an adaptive collaboration intrusion detection method to improve the safety of a network. A self-adaptive and collaborative intrusion detection model is built by applying the Environmentsclasses, agents, roles, groups, and objects(E-CARGO) model. The objects, roles, agents, and groups are designed by using decision trees(DTs) and support vector machines(SVMs), and adaptive scheduling mechanisms are set up. The KDD CUP 1999 data set is used to verify the effectiveness of the method. The experimental results demonstrate the feasibility and efficiency of the proposed collaborative and adaptive intrusion detection method. Also, the proposed method is shown to be more predominant than the methods that use a set of single type support vector machine(SVM) in terms of detection precision rate and recall rate.
基金The National Natural Science Foundation of China under contract Nos 41506198 and 41476101the Natural Science Foundation Projects of Shandong Province of China under contract No.ZR2012FZ003the Science and Technology Development Plan of Qingdao City of China under contract No.13-1-4-121-jch
文摘Since the atmospheric correction is a necessary preprocessing step of remote sensing image before detecting green tide, the introduced error directly affects the detection precision. Therefore, the detection method of green tide is presented from Landsat TM/ETM plus image which needs not the atmospheric correction. In order to achieve an automatic detection of green tide, a linear relationship（y =0.723 x＋0.504） between detection threshold y and subtraction x（x=λnir–λred） is found from the comparing Landsat TM/ETM plus image with the field surveys.Using this relationship, green tide patches can be detected automatically from Landsat TM/ETM plus image.Considering there is brightness difference between different regions in an image, the image will be divided into a plurality of windows（sub-images） with a same size firstly, and then each window will be detected using an adaptive detection threshold determined according to the discovered linear relationship. It is found that big errors will appear in some windows, such as those covered by clouds seriously. To solve this problem, the moving step k of windows is proposed to be less than the window width n. Using this mechanism, most pixels will be detected[n/k]×[n/k] times except the boundary pixels, then every pixel will be assigned the final class（green tide or sea water） according to majority rule voting strategy. It can be seen from the experiments, the proposed detection method using multi-windows and their adaptive thresholds can detect green tide from Landsat TM/ETM plus image automatically. Meanwhile, it avoids the reliance on the accurate atmospheric correction.
基金supported by 2023 Teaching Research Project of the Education Department of Anhui Province:Exploration of Optimizing Teaching Strategies for Embedded Courses in the Context of“New Engineering”(Project No.2023jyxm0460)2024 High-quality Course on Ideological and Political Education Integrated into Curriculum at Anhui University of Engineering:“Data Structures and Algorithms”(Project No.2024szyzk40)Industry-University-Research Cooperation Project of Anhui University of Engineering:“Online detection of surface quality defects in high-speed wire rod”(Project No.HX-2024-11-003).
文摘With the gradual penetration of the internet of things(IoT)into all areas of life,the scale of IoT devices shows an explosive growth trend.The era of internet of everything is coming,and the important position of IoT security is becoming increasingly prominent.Due to the large number types of IoT devices,there may be different security vulnerabilities,and unknown attack forms and virus samples are appear.In other words,large number of IoT devices,large data volumes,and various attack forms pose a big challenge of malicious traffic identification.To solve these problems,this paper proposes a concept drift detection and adaptation(CDDA)method for IoT security framework.The AI model performance is evaluated by verifying the effectiveness of IoT traffic for data drift detection,so as to select the best AI model.The experimental test are given to confirm that the feasibility of the framework and the adaptive method in practice,and the effect on the performance of IoT traffic identification is also verified.
基金Project(61701060)supported by the National Natural Science Foundation of China。
文摘Visual background extraction algorithm(ViBe)uses the first frame image to initialize the background model,which can easily introduce the“ghost”.Because ViBe uses the fixed segmentation threshold to achieve the foreground and background segmentation,the detection results in many false detections for the highly dynamic background.To solve these problems,an improved ghost suppression and adaptive Visual Background Extraction algorithm is proposed in this paper.Firstly,with the pixel’s temporal and spatial information,the historical pixels of a certain combination are used to initialize the background model in the odd frames of the video sequence.Secondly,the background sample set combined with the neighborhood pixels are used to determine a complex degree of the background,to acquire the adaptive segmentation threshold.Thirdly,the update rate is adjusted based on the complexity of the background.Finally,the detected result goes through a post-processing to achieve better detection results.The experimental results show that the improved algorithm will not only quickly suppress the“ghost”,but also have a better detection in a complex dynamic background.
基金Support by the National Natural Science Foundation of China(61174114)the Research Fund for the Doctoral Program of Higher Education in China(20120101130016)Zhejiang Provincial Science and Technology Planning Projects of China(2014C31019)
文摘In chemical process, a large number of measured and manipulated variables are highly correlated. Principal component analysis(PCA) is widely applied as a dimension reduction technique for capturing strong correlation underlying in the process measurements. However, it is difficult for PCA based fault detection results to be interpreted physically and to provide support for isolation. Some approaches incorporating process knowledge are developed, but the information is always shortage and deficient in practice. Therefore, this work proposes an adaptive partitioning PCA algorithm entirely based on operation data. The process feature space is partitioned into several sub-feature spaces. Constructed sub-block models can not only reflect the local behavior of process change, namely to grasp the intrinsic local information underlying the process changes, but also improve the fault detection and isolation through the combination of local fault detection results and reduction of smearing effect.The method is demonstrated in TE process, and the results show that the new method is much better in fault detection and isolation compared to conventional PCA method.
基金the Natural Science Foundation of Jiangsu Province(BK20200214)National Key R&D Program of China(2017YFB0403701)+5 种基金Jiangsu Province Key R&D Program(BE2019682 and BE2018667)National Natural Science Foundation of China(61605210,61675226,and 62075235)Youth Innovation Promotion Association of Chinese Academy of Sciences(2019320)Frontier Science Research Project of the Chinese Academy of Sciences(QYZDB-SSW-JSC03)Strategic Priority Research Program of the Chinese Academy of Sciences(XDB02060000)and Entrepreneurship and Innova-tion Talents in Jiangsu Province(Innovation of Scienti¯c Research Institutes).
文摘Cone photoreceptor cell identication is important for the early diagnosis of retinopathy.In this study,an object detection algorithm is used for cone cell identication in confocal adaptive optics scanning laser ophthalmoscope(AOSLO)images.An effectiveness evaluation of identication using the proposed method reveals precision,recall,and F_(1)-score of 95.8%,96.5%,and 96.1%,respectively,considering manual identication as the ground truth.Various object detection and identication results from images with different cone photoreceptor cell distributions further demonstrate the performance of the proposed method.Overall,the proposed method can accurately identify cone photoreceptor cells on confocal adaptive optics scanning laser ophthalmoscope images,being comparable to manual identication.
文摘Low Resolution Thermal Array Sensors are widely used in several applications in indoor environments. In particular, one of these cheap, small and unobtrusive sensors provides a low-resolution thermal image of the environment and, unlike cameras;it is capable to detect human heat emission even in dark rooms. The obtained thermal data can be used to monitor older seniors while they are performing daily activities at home, to detect critical situations such as falls. Most of the studies in activity recognition using Thermal Array Sensors require human detection techniques to recognize humans passing in the sensor field of view. This paper aims to improve the accuracy of the algorithms used so far by considering the temperature environment variation. This method leverages an adaptive background estimation and a noise removal technique based on Kalman Filter. In order to properly validate the system, a novel installation of a single sensor has been implemented in a smart environment: the obtained results show an improvement in human detection accuracy with respect to the state of the art, especially in case of disturbed environments.
基金The National Natural Science Foundation of China (32371993)The Natural Science Research Key Project of Anhui Provincial University(2022AH040125&2023AH040135)The Key Research and Development Plan of Anhui Province (202204c06020022&2023n06020057)。
文摘This study aimed to address the challenge of accurately and reliably detecting tomatoes in dense planting environments,a critical prerequisite for the automation implementation of robotic harvesting.However,the heavy reliance on extensive manually annotated datasets for training deep learning models still poses significant limitations to their application in real-world agricultural production environments.To overcome these limitations,we employed domain adaptive learning approach combined with the YOLOv5 model to develop a novel tomato detection model called as TDA-YOLO(tomato detection domain adaptation).We designated the normal illumination scenes in dense planting environments as the source domain and utilized various other illumination scenes as the target domain.To construct bridge mechanism between source and target domains,neural preset for color style transfer is introduced to generate a pseudo-dataset,which served to deal with domain discrepancy.Furthermore,this study combines the semi-supervised learning method to enable the model to extract domain-invariant features more fully,and uses knowledge distillation to improve the model's ability to adapt to the target domain.Additionally,for purpose of promoting inference speed and low computational demand,the lightweight FasterNet network was integrated into the YOLOv5's C3 module,creating a modified C3_Faster module.The experimental results demonstrated that the proposed TDA-YOLO model significantly outperformed original YOLOv5s model,achieving a mAP(mean average precision)of 96.80%for tomato detection across diverse scenarios in dense planting environments,increasing by 7.19 percentage points;Compared with the latest YOLOv8 and YOLOv9,it is also 2.17 and 1.19 percentage points higher,respectively.The model's average detection time per image was an impressive 15 milliseconds,with a FLOPs(floating point operations per second)count of 13.8 G.After acceleration processing,the detection accuracy of the TDA-YOLO model on the Jetson Xavier NX development board is 90.95%,the mAP value is 91.35%,and the detection time of each image is 21 ms,which can still meet the requirements of real-time detection of tomatoes in dense planting environment.The experimental results show that the proposed TDA-YOLO model can accurately and quickly detect tomatoes in dense planting environment,and at the same time avoid the use of a large number of annotated data,which provides technical support for the development of automatic harvesting systems for tomatoes and other fruits.
文摘An adaptive endpoint detection algorithm based on band energy and adaptive smoothing algorithm is described. This algorithm utilizes the capability of adaptive smoothing algorithm that intensifies the discontinuity between local areas. The band energy features are selected because of their usefulness in detecting high energy regions (in the incoming signal) and making the distinction between speech and noise. Heuristic 'edge-focusing' is used to endpoint detection to save the time in iteration.
基金supported by the National Natural Science Foundation of China(61371181)the Shandong Provincial Natural Science Foundation(ZR2012FQ007)the Natural Scientific Research Innovation Foundation in Harbin Institute of Technology(HIT.NSRIF.2011118)
文摘The detection performance and the constant false alarm rate behavior of the conventional adaptive detectors are severely degraded in heterogeneous clutter. This paper designs and analyses a knowledge-based （KB） adaptive polarimetric detector in het-erogeneous clutter. The proposed detection scheme is composed of a data selector using polarization knowledge and an adaptive polarization detector using training data. A polarization data selector based on the maximum likelihood estimation is proposed to remove outliers from the heterogeneous training data. This selector can remove outliers effectively, thus the training data is purified for estimating the clutter covariance matrix. Consequently, the performance of the adaptive detector is improved. We assess the performance of the KB adaptive polarimetric detector and the adaptive polarimetric detector without a data selector using simulated data and IPIX radar data. The results show that the KB adaptive polarization detector outperforms its non-KB counterparts.
