The development of spectroscopic survey telescopes like Large Sky Area Multi-Object Fiber Spectroscopic Telescope(LAMOST),Apache Point Observatory Galactic Evolution Experiment and Sloan Digital Sky Survey has opened ...The development of spectroscopic survey telescopes like Large Sky Area Multi-Object Fiber Spectroscopic Telescope(LAMOST),Apache Point Observatory Galactic Evolution Experiment and Sloan Digital Sky Survey has opened up unprecedented opportunities for stellar classification.Specific types of stars,such as early-type emission-line stars and those with stellar winds,can be distinguished by the profiles of their spectral lines.In this paper,we introduce a method based on derivative spectroscopy(DS)designed to detect signals within complex backgrounds and provide a preliminary estimation of curve profiles.This method exhibits a unique advantage in identifying weak signals and unusual spectral line profiles when compared to other popular line detection methods.We validated our approach using synthesis spectra,demonstrating that DS can detect emission signals three times fainter than Gaussian fitting.Furthermore,we applied our method to 579,680 co-added spectra from LAMOST Medium-Resolution Spectroscopic Survey,identifying 16,629 spectra with emission peaks around the Hαline from 10,963 stars.These spectra were classified into three distinct morphological groups,resulting in nine subclasses as follows.(1)Emission peak above the pseudo-continuum line(single peak,double peaks,emission peak situated within an absorption line,P Cygni profile,Inverse P Cygni profile);(2)Emission peak below the pseudo-continuum line(sharp emission peak,double absorption peaks,emission peak shifted to one side of the absorption line);(3)Emission peak between the pseudo-continuum line.展开更多
The advancement of autonomous driving heavily relies on the ability to accurate lane lines detection.As deep learning and computer vision technologies evolve,a variety of deep learning-based methods for lane line dete...The advancement of autonomous driving heavily relies on the ability to accurate lane lines detection.As deep learning and computer vision technologies evolve,a variety of deep learning-based methods for lane line detection have been proposed by researchers in the field.However,owing to the simple appearance of lane lines and the lack of distinctive features,it is easy for other objects with similar local appearances to interfere with the process of detecting lane lines.The precision of lane line detection is limited by the unpredictable quantity and diversity of lane lines.To address the aforementioned challenges,we propose a novel deep learning approach for lane line detection.This method leverages the Swin Transformer in conjunction with LaneNet(called ST-LaneNet).The experience results showed that the true positive detection rate can reach 97.53%for easy lanes and 96.83%for difficult lanes(such as scenes with severe occlusion and extreme lighting conditions),which can better accomplish the objective of detecting lane lines.In 1000 detection samples,the average detection accuracy can reach 97.83%,the average inference time per image can reach 17.8 ms,and the average number of frames per second can reach 64.8 Hz.The programming scripts and associated models for this project can be accessed openly at the following GitHub repository:https://github.com/Duane 711/Lane-line-detec tion-ST-LaneNet.展开更多
Diagnosing various diseases such as glaucoma,age-related macular degeneration,cardiovascular conditions,and diabetic retinopathy involves segmenting retinal blood vessels.The task is particularly challenging when deal...Diagnosing various diseases such as glaucoma,age-related macular degeneration,cardiovascular conditions,and diabetic retinopathy involves segmenting retinal blood vessels.The task is particularly challenging when dealing with color fundus images due to issues like non-uniformillumination,low contrast,and variations in vessel appearance,especially in the presence of different pathologies.Furthermore,the speed of the retinal vessel segmentation system is of utmost importance.With the surge of now available big data,the speed of the algorithm becomes increasingly important,carrying almost equivalent weightage to the accuracy of the algorithm.To address these challenges,we present a novel approach for retinal vessel segmentation,leveraging efficient and robust techniques based on multiscale line detection and mathematical morphology.Our algorithm’s performance is evaluated on two publicly available datasets,namely the Digital Retinal Images for Vessel Extraction dataset(DRIVE)and the Structure Analysis of Retina(STARE)dataset.The experimental results demonstrate the effectiveness of our method,withmean accuracy values of 0.9467 forDRIVE and 0.9535 for STARE datasets,aswell as sensitivity values of 0.6952 forDRIVE and 0.6809 for STARE datasets.Notably,our algorithmexhibits competitive performance with state-of-the-art methods.Importantly,it operates at an average speed of 3.73 s per image for DRIVE and 3.75 s for STARE datasets.It is worth noting that these results were achieved using Matlab scripts containing multiple loops.This suggests that the processing time can be further reduced by replacing loops with vectorization.Thus the proposed algorithm can be deployed in real time applications.In summary,our proposed system strikes a fine balance between swift computation and accuracy that is on par with the best available methods in the field.展开更多
Insulator defect detection plays a vital role in maintaining the secure operation of power systems.To address the issues of the difficulty of detecting small objects and missing objects due to the small scale,variable...Insulator defect detection plays a vital role in maintaining the secure operation of power systems.To address the issues of the difficulty of detecting small objects and missing objects due to the small scale,variable scale,and fuzzy edge morphology of insulator defects,we construct an insulator dataset with 1600 samples containing flashovers and breakages.Then a simple and effective surface defect detection method of power line insulators for difficult small objects is proposed.Firstly,a high-resolution featuremap is introduced and a small object prediction layer is added so that the model can detect tiny objects.Secondly,a simplified adaptive spatial feature fusion(SASFF)module is introduced to perform cross-scale spatial fusion to improve adaptability to variable multi-scale features.Finally,we propose an enhanced deformable attention mechanism(EDAM)module.By integrating a gating activation function,the model is further inspired to learn a small number of critical sampling points near reference points.And the module can improve the perception of object morphology.The experimental results indicate that concerning the dataset of flashover and breakage defects,this method improves the performance of YOLOv5,YOLOv7,and YOLOv8.In practical application,it can simply and effectively improve the precision of power line insulator defect detection and reduce missing detection for difficult small objects.展开更多
The transmission line tower will be affected by bad weather and artificial subsidence caused by the foundation and other factors in the power transmission.The tower’s tilt and severe deformation will cause the buildi...The transmission line tower will be affected by bad weather and artificial subsidence caused by the foundation and other factors in the power transmission.The tower’s tilt and severe deformation will cause the building to collapse.Many small changes caused the tower’s collapse,but the early staff often could not intuitively notice the changes in the tower’s state.In the current tower online monitoring system,terminal equipment often needs to replace batteries frequently due to premature exhaustion of power.According to the need for real-time measurement of power line tower,this research designed a real-time monitoring device monitoring the transmission tower attitude tilting and foundation state based on the inertial sensor,the acceleration of 3 axis inertial sensor and angular velocity raw data to pole average filtering pre-processing,and then through the complementary filtering algorithm for comprehensive calculation of tilt angle,the system meets the demand for inclined online monitoring of power line poles and towers regarding measurement accuracy,with low cost and power consumption.The optimization multi-sensor cooperative detection and correction measured tilt angle result relative accuracy can reach 1.03%,which has specific promotion and application value since the system has the advantages of unattended and efficient calculation.展开更多
Wearing helmetswhile riding electric bicycles can significantly reduce head injuries resulting fromtraffic accidents.To effectively monitor compliance,the utilization of target detection algorithms through traffic cam...Wearing helmetswhile riding electric bicycles can significantly reduce head injuries resulting fromtraffic accidents.To effectively monitor compliance,the utilization of target detection algorithms through traffic cameras plays a vital role in identifying helmet usage by electric bicycle riders and recognizing license plates on electric bicycles.However,manual enforcement by traffic police is time-consuming and labor-intensive.Traditional methods face challenges in accurately identifying small targets such as helmets and license plates using deep learning techniques.This paper proposes an enhanced model for detecting helmets and license plates on electric bicycles,addressing these challenges.The proposedmodel improves uponYOLOv8n by deepening the network structure,incorporating weighted connections,and introducing lightweight convolutional modules.These modifications aim to enhance the precision of small target recognition while reducing the model’s parameters,making it suitable for deployment on low-performance devices in real traffic scenarios.Experimental results demonstrate that the model achieves an mAP@0.5 of 91.8%,showing an 11.5%improvement over the baselinemodel,with a 16.2%reduction in parameters.Additionally,themodel achieves a frames per second(FPS)rate of 58,meeting the accuracy and speed requirements for detection in actual traffic scenarios.展开更多
Urban underground pipelines are an important infrastructure in cities,and timely investigation of problems in underground pipelines can help ensure the normal operation of cities.Owing to the growing demand for defect...Urban underground pipelines are an important infrastructure in cities,and timely investigation of problems in underground pipelines can help ensure the normal operation of cities.Owing to the growing demand for defect detection in urban underground pipelines,this study developed an improved defect detection method for urban underground pipelines based on fully convolutional one-stage object detector(FCOS),called spatial pyramid pooling-fast(SPPF)feature fusion and dual detection heads based on FCOS(SDH-FCOS)model.This study improved the feature fusion component of the model network based on FCOS,introduced an SPPF network structure behind the last output feature layer of the backbone network,fused the local and global features,added a top-down path to accelerate the circulation of shallowinformation,and enriched the semantic information acquired by shallow features.The ability of the model to detect objects with multiple morphologies was strengthened by introducing dual detection heads.The experimental results using an open dataset of underground pipes show that the proposed SDH-FCOS model can recognize underground pipe defects more accurately;the average accuracy was improved by 2.7% compared with the original FCOS model,reducing the leakage rate to a large extent and achieving real-time detection.Also,our model achieved a good trade-off between accuracy and speed compared with other mainstream methods.This proved the effectiveness of the proposed model.展开更多
Data security assurance is crucial due to the increasing prevalence of cloud computing and its widespread use across different industries,especially in light of the growing number of cybersecurity threats.A major and ...Data security assurance is crucial due to the increasing prevalence of cloud computing and its widespread use across different industries,especially in light of the growing number of cybersecurity threats.A major and everpresent threat is Ransomware-as-a-Service(RaaS)assaults,which enable even individuals with minimal technical knowledge to conduct ransomware operations.This study provides a new approach for RaaS attack detection which uses an ensemble of deep learning models.For this purpose,the network intrusion detection dataset“UNSWNB15”from the Intelligent Security Group of the University of New South Wales,Australia is analyzed.In the initial phase,the rectified linear unit-,scaled exponential linear unit-,and exponential linear unit-based three separate Multi-Layer Perceptron(MLP)models are developed.Later,using the combined predictive power of these three MLPs,the RansoDetect Fusion ensemble model is introduced in the suggested methodology.The proposed ensemble technique outperforms previous studieswith impressive performance metrics results,including 98.79%accuracy and recall,98.85%precision,and 98.80%F1-score.The empirical results of this study validate the ensemble model’s ability to improve cybersecurity defenses by showing that it outperforms individual MLPmodels.In expanding the field of cybersecurity strategy,this research highlights the significance of combined deep learning models in strengthening intrusion detection systems against sophisticated cyber threats.展开更多
In the era of the Internet,widely used web applications have become the target of hacker attacks because they contain a large amount of personal information.Among these vulnerabilities,stealing private data through cr...In the era of the Internet,widely used web applications have become the target of hacker attacks because they contain a large amount of personal information.Among these vulnerabilities,stealing private data through crosssite scripting(XSS)attacks is one of the most commonly used attacks by hackers.Currently,deep learning-based XSS attack detection methods have good application prospects;however,they suffer from problems such as being prone to overfitting,a high false alarm rate,and low accuracy.To address these issues,we propose a multi-stage feature extraction and fusion model for XSS detection based on Random Forest feature enhancement.The model utilizes RandomForests to capture the intrinsic structure and patterns of the data by extracting leaf node indices as features,which are subsequentlymergedwith the original data features to forma feature setwith richer information content.Further feature extraction is conducted through three parallel channels.Channel I utilizes parallel onedimensional convolutional layers(1Dconvolutional layers)with different convolutional kernel sizes to extract local features at different scales and performmulti-scale feature fusion;Channel II employsmaximum one-dimensional pooling layers(max 1D pooling layers)of various sizes to extract key features from the data;and Channel III extracts global information bi-directionally using a Bi-Directional Long-Short TermMemory Network(Bi-LSTM)and incorporates a multi-head attention mechanism to enhance global features.Finally,effective classification and prediction of XSS are performed by fusing the features of the three channels.To test the effectiveness of the model,we conduct experiments on six datasets.We achieve an accuracy of 100%on the UNSW-NB15 dataset and 99.99%on the CICIDS2017 dataset,which is higher than that of the existing models.展开更多
With the advent of Industry 4.0,marked by a surge in intelligent manufacturing,advanced sensors embedded in smart factories now enable extensive data collection on equipment operation.The analysis of such data is pivo...With the advent of Industry 4.0,marked by a surge in intelligent manufacturing,advanced sensors embedded in smart factories now enable extensive data collection on equipment operation.The analysis of such data is pivotal for ensuring production safety,a critical factor in monitoring the health status of manufacturing apparatus.Conventional defect detection techniques,typically limited to specific scenarios,often require manual feature extraction,leading to inefficiencies and limited versatility in the overall process.Our research presents an intelligent defect detection methodology that leverages deep learning techniques to automate feature extraction and defect localization processes.Our proposed approach encompasses a suite of components:the high-level feature learning block(HLFLB),the multi-scale feature learning block(MSFLB),and a dynamic adaptive fusion block(DAFB),working in tandem to extract meticulously and synergistically aggregate defect-related characteristics across various scales and hierarchical levels.We have conducted validation of the proposed method using datasets derived from gearbox and bearing assessments.The empirical outcomes underscore the superior defect detection capability of our approach.It demonstrates consistently high performance across diverse datasets and possesses the accuracy required to categorize defects,taking into account their specific locations and the extent of damage,proving the method’s effectiveness and reliability in identifying defects in industrial components.展开更多
Online review platforms are becoming increasingly popular,encouraging dishonest merchants and service providers to deceive customers by creating fake reviews for their goods or services.Using Sybil accounts,bot farms,...Online review platforms are becoming increasingly popular,encouraging dishonest merchants and service providers to deceive customers by creating fake reviews for their goods or services.Using Sybil accounts,bot farms,and real account purchases,immoral actors demonize rivals and advertise their goods.Most academic and industry efforts have been aimed at detecting fake/fraudulent product or service evaluations for years.The primary hurdle to identifying fraudulent reviews is the lack of a reliable means to distinguish fraudulent reviews from real ones.This paper adopts a semi-supervised machine learning method to detect fake reviews on any website,among other things.Online reviews are classified using a semi-supervised approach(PU-learning)since there is a shortage of labeled data,and they are dynamic.Then,classification is performed using the machine learning techniques Support Vector Machine(SVM)and Nave Bayes.The performance of the suggested system has been compared with standard works,and experimental findings are assessed using several assessment metrics.展开更多
Aiming at the challenges associated with the absence of a labeled dataset for Yi characters and the complexity of Yi character detection and recognition,we present a deep learning-based approach for Yi character detec...Aiming at the challenges associated with the absence of a labeled dataset for Yi characters and the complexity of Yi character detection and recognition,we present a deep learning-based approach for Yi character detection and recognition.In the detection stage,an improved Differentiable Binarization Network(DBNet)framework is introduced to detect Yi characters,in which the Omni-dimensional Dynamic Convolution(ODConv)is combined with the ResNet-18 feature extraction module to obtain multi-dimensional complementary features,thereby improving the accuracy of Yi character detection.Then,the feature pyramid network fusion module is used to further extract Yi character image features,improving target recognition at different scales.Further,the previously generated feature map is passed through a head network to produce two maps:a probability map and an adaptive threshold map of the same size as the original map.These maps are then subjected to a differentiable binarization process,resulting in an approximate binarization map.This map helps to identify the boundaries of the text boxes.Finally,the text detection box is generated after the post-processing stage.In the recognition stage,an improved lightweight MobileNetV3 framework is used to recognize the detect character regions,where the original Squeeze-and-Excitation(SE)block is replaced by the efficient Shuffle Attention(SA)that integrates spatial and channel attention,improving the accuracy of Yi characters recognition.Meanwhile,the use of depth separable convolution and reversible residual structure can reduce the number of parameters and computation of the model,so that the model can better understand the contextual information and improve the accuracy of text recognition.The experimental results illustrate that the proposed method achieves good results in detecting and recognizing Yi characters,with detection and recognition accuracy rates of 97.5%and 96.8%,respectively.And also,we have compared the detection and recognition algorithms proposed in this paper with other typical algorithms.In these comparisons,the proposed model achieves better detection and recognition results with a certain reliability.展开更多
Infrared signal detection is widely used in many fields.Due to the detection principle,however,the accuracy and range of detection are limited.Thanks to the ultra stability of the^(87)Sr optical lattice clock,external...Infrared signal detection is widely used in many fields.Due to the detection principle,however,the accuracy and range of detection are limited.Thanks to the ultra stability of the^(87)Sr optical lattice clock,external infrared electromagnetic wave disturbances can be responded to.Utilizing the ac Stark shift of the clock transition,we propose a new method to detect infrared signals.According to our calculations,the theoretical detection accuracy in the vicinity of its resonance band of 2.6μm can reach the order of 10-14W,while the minimum detectable signal of common detectors is on the order of 10^(-10)W.展开更多
Detecting a pipeline's abnormal status,which is typically a blockage and leakage accident,is important for the continuity and safety of mine backfill.The pipeline system for gravity-transport high-density backfill...Detecting a pipeline's abnormal status,which is typically a blockage and leakage accident,is important for the continuity and safety of mine backfill.The pipeline system for gravity-transport high-density backfill(GHB)is complex.Specifically designed,efficient,and accurate abnormal pipeline detection methods for GHB are rare.This work presents a long short-term memory-based deep learning(LSTM-DL)model for GHB pipeline blockage and leakage diagnosis.First,an industrial pipeline monitoring system was introduced using pressure and flow sensors.Second,blockage and leakage field experiments were designed to solve the problem of negative sample deficiency.The pipeline's statistical characteristics with different working statuses were analyzed to show their complexity.Third,the architecture of the LSTM-DL model was elaborated on and evaluated.Finally,the LSTM-DL model was compared with state-of-the-art(SOTA)learning algorithms.The results show that the backfilling cycle comprises multiple working phases and is intermittent.Although pressure and flow signals fluctuate stably in a normal cycle,their values are diverse in different cycles.Plugging causes a sudden change in interval signal features;leakage results in long variation duration and a wide fluctuation range.Among the SOTA models,the LSTM-DL model has the highest detection accuracy of98.31%for all states and the lowest misjudgment or false positive rate of 3.21%for blockage and leakage states.The proposed model can accurately recognize various pipeline statuses of complex GHB systems.展开更多
The continuous operation of On-Load Tap-Changers (OLTC) is essential for maintaining stable voltage levels in power transmission and distribution systems. Timely fault detection in OLTC is essential for preventing maj...The continuous operation of On-Load Tap-Changers (OLTC) is essential for maintaining stable voltage levels in power transmission and distribution systems. Timely fault detection in OLTC is essential for preventing major failures and ensuring the reliability of the electrical grid. This research paper proposes an innovative approach that combines voiceprint detection using MATLAB analysis for online fault monitoring of OLTC. By leveraging advanced signal processing techniques and machine learning algorithms in MATLAB, the proposed method accurately detects faults in OLTC, providing real-time monitoring and proactive maintenance strategies.展开更多
With the continuous integration of new energy into the power grid,various new attacks continue to emerge and the feature distributions are constantly changing during the deployment of intelligent pumped storage power ...With the continuous integration of new energy into the power grid,various new attacks continue to emerge and the feature distributions are constantly changing during the deployment of intelligent pumped storage power stations.The intrusion detection model trained on the old data is hard to effectively identify new attacks,and it is difficult to update the intrusion detection model in time when lacking data.To solve this issue,by using model-based transfer learning methods,in this paper we propose a convolutional neural network(CNN)based transfer online sequential extreme learning machine(TOS-ELM)scheme to enable the online intrusion detection,which is called CNN-TOSELM in this paper.In our proposed scheme,we use pre-trained CNN to extract the characteristics of the target domain data as input,and then build online learning classifier TOS-ELM to transfer the parameter of the ELM classifier of the source domain.Experimental results show the proposed CNNTOSELM scheme can achieve better detection performance and extremely short model update time for intelligent pumped storage power stations.展开更多
Nonlinear optical imaging is a versatile tool that has been proven to be exceptionally useful in various research fields.However,due to the use of photomultiplier tubes(PMTs),the wide application of nonlinear optical ...Nonlinear optical imaging is a versatile tool that has been proven to be exceptionally useful in various research fields.However,due to the use of photomultiplier tubes(PMTs),the wide application of nonlinear optical imaging is limited by the incapability of imaging under am-bient light.In this paper,we propose and demonstrate a new optical imaging detection method based on optical parametric amplification(OPA).As a nonlinear optical process,OPA in-trinsically rejects ambient light photons by coherence gating.Periodical poled lithium niobate(PPLN)crystals are used in this study as the media for OPA.Compared to bulk nonlinear optical crystals,PPLN crystals support the generation of OPA signal with lower pump power.Therefore,this characteristic of PPLN crystals is particularly beneficial when using high-repetition-rate lasers,which facilitate high-speed optical signal detection,such as in spec-troscopy and imaging.A PPLN-based OPA system was built to amplify the emitted imaging signal from second harmonic generation(SHG)and coherent anti-Stokes Raman scattering(CARS)microscopy imaging,and the amplified optical signal was strong enough to be detected by a biased photodiode under ordinary room light conditions.With OPA detection,ambient-light-on SHG and CARS imaging becomes possible,and achieves a similar result as PMT detection under strictly dark environments.These results demonstrate that OPA can be used as a substitute for PMTs in nonlinear optical imaging to adapt it to various applications with complex.light ing conditions.展开更多
The Cloud system shows its growing functionalities in various industrial applications.The safety towards data transfer seems to be a threat where Network Intrusion Detection System(NIDS)is measured as an essential ele...The Cloud system shows its growing functionalities in various industrial applications.The safety towards data transfer seems to be a threat where Network Intrusion Detection System(NIDS)is measured as an essential element to fulfill security.Recently,Machine Learning(ML)approaches have been used for the construction of intellectual IDS.Most IDS are based on ML techniques either as unsupervised or supervised.In supervised learning,NIDS is based on labeled data where it reduces the efficiency of the reduced model to identify attack patterns.Similarly,the unsupervised model fails to provide a satisfactory outcome.Hence,to boost the functionality of unsupervised learning,an effectual auto-encoder is applied for feature selection to select good features.Finally,the Naïve Bayes classifier is used for classification purposes.This approach exposes the finest generalization ability to train the data.The unlabelled data is also used for adoption towards data analysis.Here,redundant and noisy samples over the dataset are eliminated.To validate the robustness and efficiency of NIDS,the anticipated model is tested over the NSL-KDD dataset.The experimental outcomes demonstrate that the anticipated approach attains superior accuracy with 93%,which is higher compared to J48,AB tree,Random Forest(RF),Regression Tree(RT),Multi-Layer Perceptrons(MLP),Support Vector Machine(SVM),and Fuzzy.Similarly,False Alarm Rate(FAR)and True Positive Rate(TPR)of Naive Bayes(NB)is 0.3 and 0.99,respectively.When compared to prevailing techniques,the anticipated approach also delivers promising outcomes.展开更多
基金the support provided by the National Natural Science Foundation of China(NSFC,Grant Nos.12090040/3,12125303,12288102,and 11733008)the National Key Research and Development Program of China(grant No.2021YFA1600401/3)+3 种基金the China Manned Space Project(CMSCSST-2021-A10)the Yunnan Fundamental Research Projects(grant No.202101AV070001)the National Natural Science Foundation of China and the Chinese Academy of Sciences,under grant No.U1831125the Research Program of Frontier Sciences,CAS(grant No.QYZDY-SSW-SLH007)。
文摘The development of spectroscopic survey telescopes like Large Sky Area Multi-Object Fiber Spectroscopic Telescope(LAMOST),Apache Point Observatory Galactic Evolution Experiment and Sloan Digital Sky Survey has opened up unprecedented opportunities for stellar classification.Specific types of stars,such as early-type emission-line stars and those with stellar winds,can be distinguished by the profiles of their spectral lines.In this paper,we introduce a method based on derivative spectroscopy(DS)designed to detect signals within complex backgrounds and provide a preliminary estimation of curve profiles.This method exhibits a unique advantage in identifying weak signals and unusual spectral line profiles when compared to other popular line detection methods.We validated our approach using synthesis spectra,demonstrating that DS can detect emission signals three times fainter than Gaussian fitting.Furthermore,we applied our method to 579,680 co-added spectra from LAMOST Medium-Resolution Spectroscopic Survey,identifying 16,629 spectra with emission peaks around the Hαline from 10,963 stars.These spectra were classified into three distinct morphological groups,resulting in nine subclasses as follows.(1)Emission peak above the pseudo-continuum line(single peak,double peaks,emission peak situated within an absorption line,P Cygni profile,Inverse P Cygni profile);(2)Emission peak below the pseudo-continuum line(sharp emission peak,double absorption peaks,emission peak shifted to one side of the absorption line);(3)Emission peak between the pseudo-continuum line.
基金Supported by National Natural Science Foundation of China(Grant Nos.51605003,51575001)Natural Science Foundation of Anhui Higher Education Institutions of China(Grant No.KJ2020A0358)Young and Middle-Aged Top Talents Training Program of Anhui Polytechnic University of China.
文摘The advancement of autonomous driving heavily relies on the ability to accurate lane lines detection.As deep learning and computer vision technologies evolve,a variety of deep learning-based methods for lane line detection have been proposed by researchers in the field.However,owing to the simple appearance of lane lines and the lack of distinctive features,it is easy for other objects with similar local appearances to interfere with the process of detecting lane lines.The precision of lane line detection is limited by the unpredictable quantity and diversity of lane lines.To address the aforementioned challenges,we propose a novel deep learning approach for lane line detection.This method leverages the Swin Transformer in conjunction with LaneNet(called ST-LaneNet).The experience results showed that the true positive detection rate can reach 97.53%for easy lanes and 96.83%for difficult lanes(such as scenes with severe occlusion and extreme lighting conditions),which can better accomplish the objective of detecting lane lines.In 1000 detection samples,the average detection accuracy can reach 97.83%,the average inference time per image can reach 17.8 ms,and the average number of frames per second can reach 64.8 Hz.The programming scripts and associated models for this project can be accessed openly at the following GitHub repository:https://github.com/Duane 711/Lane-line-detec tion-ST-LaneNet.
文摘Diagnosing various diseases such as glaucoma,age-related macular degeneration,cardiovascular conditions,and diabetic retinopathy involves segmenting retinal blood vessels.The task is particularly challenging when dealing with color fundus images due to issues like non-uniformillumination,low contrast,and variations in vessel appearance,especially in the presence of different pathologies.Furthermore,the speed of the retinal vessel segmentation system is of utmost importance.With the surge of now available big data,the speed of the algorithm becomes increasingly important,carrying almost equivalent weightage to the accuracy of the algorithm.To address these challenges,we present a novel approach for retinal vessel segmentation,leveraging efficient and robust techniques based on multiscale line detection and mathematical morphology.Our algorithm’s performance is evaluated on two publicly available datasets,namely the Digital Retinal Images for Vessel Extraction dataset(DRIVE)and the Structure Analysis of Retina(STARE)dataset.The experimental results demonstrate the effectiveness of our method,withmean accuracy values of 0.9467 forDRIVE and 0.9535 for STARE datasets,aswell as sensitivity values of 0.6952 forDRIVE and 0.6809 for STARE datasets.Notably,our algorithmexhibits competitive performance with state-of-the-art methods.Importantly,it operates at an average speed of 3.73 s per image for DRIVE and 3.75 s for STARE datasets.It is worth noting that these results were achieved using Matlab scripts containing multiple loops.This suggests that the processing time can be further reduced by replacing loops with vectorization.Thus the proposed algorithm can be deployed in real time applications.In summary,our proposed system strikes a fine balance between swift computation and accuracy that is on par with the best available methods in the field.
基金State Grid Jiangsu Electric Power Co.,Ltd.of the Science and Technology Project(Grant No.J2022004).
文摘Insulator defect detection plays a vital role in maintaining the secure operation of power systems.To address the issues of the difficulty of detecting small objects and missing objects due to the small scale,variable scale,and fuzzy edge morphology of insulator defects,we construct an insulator dataset with 1600 samples containing flashovers and breakages.Then a simple and effective surface defect detection method of power line insulators for difficult small objects is proposed.Firstly,a high-resolution featuremap is introduced and a small object prediction layer is added so that the model can detect tiny objects.Secondly,a simplified adaptive spatial feature fusion(SASFF)module is introduced to perform cross-scale spatial fusion to improve adaptability to variable multi-scale features.Finally,we propose an enhanced deformable attention mechanism(EDAM)module.By integrating a gating activation function,the model is further inspired to learn a small number of critical sampling points near reference points.And the module can improve the perception of object morphology.The experimental results indicate that concerning the dataset of flashover and breakage defects,this method improves the performance of YOLOv5,YOLOv7,and YOLOv8.In practical application,it can simply and effectively improve the precision of power line insulator defect detection and reduce missing detection for difficult small objects.
基金This work was supported by the National Natural Science Foundation of China(Nos.62172242,51901152)Industry University Cooperation Education Program of the Ministry of Education(No.2020021680113)Shanxi Scholarship Council of China.
文摘The transmission line tower will be affected by bad weather and artificial subsidence caused by the foundation and other factors in the power transmission.The tower’s tilt and severe deformation will cause the building to collapse.Many small changes caused the tower’s collapse,but the early staff often could not intuitively notice the changes in the tower’s state.In the current tower online monitoring system,terminal equipment often needs to replace batteries frequently due to premature exhaustion of power.According to the need for real-time measurement of power line tower,this research designed a real-time monitoring device monitoring the transmission tower attitude tilting and foundation state based on the inertial sensor,the acceleration of 3 axis inertial sensor and angular velocity raw data to pole average filtering pre-processing,and then through the complementary filtering algorithm for comprehensive calculation of tilt angle,the system meets the demand for inclined online monitoring of power line poles and towers regarding measurement accuracy,with low cost and power consumption.The optimization multi-sensor cooperative detection and correction measured tilt angle result relative accuracy can reach 1.03%,which has specific promotion and application value since the system has the advantages of unattended and efficient calculation.
基金supported by the Ningxia Key Research and Development Program(Talent Introduction Special Project)Project(2022YCZX0013)North Minzu University 2022 School-Level Scientific Research Platform“Digital Agriculture Enabling Ningxia Rural Revitalization Innovation Team”(2022PT_S10)+1 种基金Yinchuan City University-Enterprise Joint Innovation Project(2022XQZD009)Ningxia Key Research and Development Program(Key Project)Project(2023BDE02001).
文摘Wearing helmetswhile riding electric bicycles can significantly reduce head injuries resulting fromtraffic accidents.To effectively monitor compliance,the utilization of target detection algorithms through traffic cameras plays a vital role in identifying helmet usage by electric bicycle riders and recognizing license plates on electric bicycles.However,manual enforcement by traffic police is time-consuming and labor-intensive.Traditional methods face challenges in accurately identifying small targets such as helmets and license plates using deep learning techniques.This paper proposes an enhanced model for detecting helmets and license plates on electric bicycles,addressing these challenges.The proposedmodel improves uponYOLOv8n by deepening the network structure,incorporating weighted connections,and introducing lightweight convolutional modules.These modifications aim to enhance the precision of small target recognition while reducing the model’s parameters,making it suitable for deployment on low-performance devices in real traffic scenarios.Experimental results demonstrate that the model achieves an mAP@0.5 of 91.8%,showing an 11.5%improvement over the baselinemodel,with a 16.2%reduction in parameters.Additionally,themodel achieves a frames per second(FPS)rate of 58,meeting the accuracy and speed requirements for detection in actual traffic scenarios.
基金supported by the National Natural Science Foundation of China under Grant No.61976226the Research and Academic Team of South-CentralMinzu University under Grant No.KTZ20050.
文摘Urban underground pipelines are an important infrastructure in cities,and timely investigation of problems in underground pipelines can help ensure the normal operation of cities.Owing to the growing demand for defect detection in urban underground pipelines,this study developed an improved defect detection method for urban underground pipelines based on fully convolutional one-stage object detector(FCOS),called spatial pyramid pooling-fast(SPPF)feature fusion and dual detection heads based on FCOS(SDH-FCOS)model.This study improved the feature fusion component of the model network based on FCOS,introduced an SPPF network structure behind the last output feature layer of the backbone network,fused the local and global features,added a top-down path to accelerate the circulation of shallowinformation,and enriched the semantic information acquired by shallow features.The ability of the model to detect objects with multiple morphologies was strengthened by introducing dual detection heads.The experimental results using an open dataset of underground pipes show that the proposed SDH-FCOS model can recognize underground pipe defects more accurately;the average accuracy was improved by 2.7% compared with the original FCOS model,reducing the leakage rate to a large extent and achieving real-time detection.Also,our model achieved a good trade-off between accuracy and speed compared with other mainstream methods.This proved the effectiveness of the proposed model.
基金the Deanship of Scientific Research,Najran University,Kingdom of Saudi Arabia,for funding this work under the Research Groups Funding Program Grant Code Number(NU/RG/SERC/12/43).
文摘Data security assurance is crucial due to the increasing prevalence of cloud computing and its widespread use across different industries,especially in light of the growing number of cybersecurity threats.A major and everpresent threat is Ransomware-as-a-Service(RaaS)assaults,which enable even individuals with minimal technical knowledge to conduct ransomware operations.This study provides a new approach for RaaS attack detection which uses an ensemble of deep learning models.For this purpose,the network intrusion detection dataset“UNSWNB15”from the Intelligent Security Group of the University of New South Wales,Australia is analyzed.In the initial phase,the rectified linear unit-,scaled exponential linear unit-,and exponential linear unit-based three separate Multi-Layer Perceptron(MLP)models are developed.Later,using the combined predictive power of these three MLPs,the RansoDetect Fusion ensemble model is introduced in the suggested methodology.The proposed ensemble technique outperforms previous studieswith impressive performance metrics results,including 98.79%accuracy and recall,98.85%precision,and 98.80%F1-score.The empirical results of this study validate the ensemble model’s ability to improve cybersecurity defenses by showing that it outperforms individual MLPmodels.In expanding the field of cybersecurity strategy,this research highlights the significance of combined deep learning models in strengthening intrusion detection systems against sophisticated cyber threats.
文摘In the era of the Internet,widely used web applications have become the target of hacker attacks because they contain a large amount of personal information.Among these vulnerabilities,stealing private data through crosssite scripting(XSS)attacks is one of the most commonly used attacks by hackers.Currently,deep learning-based XSS attack detection methods have good application prospects;however,they suffer from problems such as being prone to overfitting,a high false alarm rate,and low accuracy.To address these issues,we propose a multi-stage feature extraction and fusion model for XSS detection based on Random Forest feature enhancement.The model utilizes RandomForests to capture the intrinsic structure and patterns of the data by extracting leaf node indices as features,which are subsequentlymergedwith the original data features to forma feature setwith richer information content.Further feature extraction is conducted through three parallel channels.Channel I utilizes parallel onedimensional convolutional layers(1Dconvolutional layers)with different convolutional kernel sizes to extract local features at different scales and performmulti-scale feature fusion;Channel II employsmaximum one-dimensional pooling layers(max 1D pooling layers)of various sizes to extract key features from the data;and Channel III extracts global information bi-directionally using a Bi-Directional Long-Short TermMemory Network(Bi-LSTM)and incorporates a multi-head attention mechanism to enhance global features.Finally,effective classification and prediction of XSS are performed by fusing the features of the three channels.To test the effectiveness of the model,we conduct experiments on six datasets.We achieve an accuracy of 100%on the UNSW-NB15 dataset and 99.99%on the CICIDS2017 dataset,which is higher than that of the existing models.
基金supported by the Natural Science Foundation of Heilongjiang Province(Grant Number:LH2021F002).
文摘With the advent of Industry 4.0,marked by a surge in intelligent manufacturing,advanced sensors embedded in smart factories now enable extensive data collection on equipment operation.The analysis of such data is pivotal for ensuring production safety,a critical factor in monitoring the health status of manufacturing apparatus.Conventional defect detection techniques,typically limited to specific scenarios,often require manual feature extraction,leading to inefficiencies and limited versatility in the overall process.Our research presents an intelligent defect detection methodology that leverages deep learning techniques to automate feature extraction and defect localization processes.Our proposed approach encompasses a suite of components:the high-level feature learning block(HLFLB),the multi-scale feature learning block(MSFLB),and a dynamic adaptive fusion block(DAFB),working in tandem to extract meticulously and synergistically aggregate defect-related characteristics across various scales and hierarchical levels.We have conducted validation of the proposed method using datasets derived from gearbox and bearing assessments.The empirical outcomes underscore the superior defect detection capability of our approach.It demonstrates consistently high performance across diverse datasets and possesses the accuracy required to categorize defects,taking into account their specific locations and the extent of damage,proving the method’s effectiveness and reliability in identifying defects in industrial components.
文摘Online review platforms are becoming increasingly popular,encouraging dishonest merchants and service providers to deceive customers by creating fake reviews for their goods or services.Using Sybil accounts,bot farms,and real account purchases,immoral actors demonize rivals and advertise their goods.Most academic and industry efforts have been aimed at detecting fake/fraudulent product or service evaluations for years.The primary hurdle to identifying fraudulent reviews is the lack of a reliable means to distinguish fraudulent reviews from real ones.This paper adopts a semi-supervised machine learning method to detect fake reviews on any website,among other things.Online reviews are classified using a semi-supervised approach(PU-learning)since there is a shortage of labeled data,and they are dynamic.Then,classification is performed using the machine learning techniques Support Vector Machine(SVM)and Nave Bayes.The performance of the suggested system has been compared with standard works,and experimental findings are assessed using several assessment metrics.
基金The work was supported by the National Natural Science Foundation of China(61972062,62306060)the Basic Research Project of Liaoning Province(2023JH2/101300191)+1 种基金the Liaoning Doctoral Research Start-Up Fund Project(2023-BS-078)the Dalian Academy of Social Sciences(2023dlsky028).
文摘Aiming at the challenges associated with the absence of a labeled dataset for Yi characters and the complexity of Yi character detection and recognition,we present a deep learning-based approach for Yi character detection and recognition.In the detection stage,an improved Differentiable Binarization Network(DBNet)framework is introduced to detect Yi characters,in which the Omni-dimensional Dynamic Convolution(ODConv)is combined with the ResNet-18 feature extraction module to obtain multi-dimensional complementary features,thereby improving the accuracy of Yi character detection.Then,the feature pyramid network fusion module is used to further extract Yi character image features,improving target recognition at different scales.Further,the previously generated feature map is passed through a head network to produce two maps:a probability map and an adaptive threshold map of the same size as the original map.These maps are then subjected to a differentiable binarization process,resulting in an approximate binarization map.This map helps to identify the boundaries of the text boxes.Finally,the text detection box is generated after the post-processing stage.In the recognition stage,an improved lightweight MobileNetV3 framework is used to recognize the detect character regions,where the original Squeeze-and-Excitation(SE)block is replaced by the efficient Shuffle Attention(SA)that integrates spatial and channel attention,improving the accuracy of Yi characters recognition.Meanwhile,the use of depth separable convolution and reversible residual structure can reduce the number of parameters and computation of the model,so that the model can better understand the contextual information and improve the accuracy of text recognition.The experimental results illustrate that the proposed method achieves good results in detecting and recognizing Yi characters,with detection and recognition accuracy rates of 97.5%and 96.8%,respectively.And also,we have compared the detection and recognition algorithms proposed in this paper with other typical algorithms.In these comparisons,the proposed model achieves better detection and recognition results with a certain reliability.
基金Project supported by the National Natural Science Foundation of China (Grant No.12274045)。
文摘Infrared signal detection is widely used in many fields.Due to the detection principle,however,the accuracy and range of detection are limited.Thanks to the ultra stability of the^(87)Sr optical lattice clock,external infrared electromagnetic wave disturbances can be responded to.Utilizing the ac Stark shift of the clock transition,we propose a new method to detect infrared signals.According to our calculations,the theoretical detection accuracy in the vicinity of its resonance band of 2.6μm can reach the order of 10-14W,while the minimum detectable signal of common detectors is on the order of 10^(-10)W.
基金financially supported by the China Postdoctoral Science Foundation (No.2021M690362)the National Natural Science Foundation of China (Nos.51974014 and U2034206)。
文摘Detecting a pipeline's abnormal status,which is typically a blockage and leakage accident,is important for the continuity and safety of mine backfill.The pipeline system for gravity-transport high-density backfill(GHB)is complex.Specifically designed,efficient,and accurate abnormal pipeline detection methods for GHB are rare.This work presents a long short-term memory-based deep learning(LSTM-DL)model for GHB pipeline blockage and leakage diagnosis.First,an industrial pipeline monitoring system was introduced using pressure and flow sensors.Second,blockage and leakage field experiments were designed to solve the problem of negative sample deficiency.The pipeline's statistical characteristics with different working statuses were analyzed to show their complexity.Third,the architecture of the LSTM-DL model was elaborated on and evaluated.Finally,the LSTM-DL model was compared with state-of-the-art(SOTA)learning algorithms.The results show that the backfilling cycle comprises multiple working phases and is intermittent.Although pressure and flow signals fluctuate stably in a normal cycle,their values are diverse in different cycles.Plugging causes a sudden change in interval signal features;leakage results in long variation duration and a wide fluctuation range.Among the SOTA models,the LSTM-DL model has the highest detection accuracy of98.31%for all states and the lowest misjudgment or false positive rate of 3.21%for blockage and leakage states.The proposed model can accurately recognize various pipeline statuses of complex GHB systems.
文摘The continuous operation of On-Load Tap-Changers (OLTC) is essential for maintaining stable voltage levels in power transmission and distribution systems. Timely fault detection in OLTC is essential for preventing major failures and ensuring the reliability of the electrical grid. This research paper proposes an innovative approach that combines voiceprint detection using MATLAB analysis for online fault monitoring of OLTC. By leveraging advanced signal processing techniques and machine learning algorithms in MATLAB, the proposed method accurately detects faults in OLTC, providing real-time monitoring and proactive maintenance strategies.
文摘With the continuous integration of new energy into the power grid,various new attacks continue to emerge and the feature distributions are constantly changing during the deployment of intelligent pumped storage power stations.The intrusion detection model trained on the old data is hard to effectively identify new attacks,and it is difficult to update the intrusion detection model in time when lacking data.To solve this issue,by using model-based transfer learning methods,in this paper we propose a convolutional neural network(CNN)based transfer online sequential extreme learning machine(TOS-ELM)scheme to enable the online intrusion detection,which is called CNN-TOSELM in this paper.In our proposed scheme,we use pre-trained CNN to extract the characteristics of the target domain data as input,and then build online learning classifier TOS-ELM to transfer the parameter of the ELM classifier of the source domain.Experimental results show the proposed CNNTOSELM scheme can achieve better detection performance and extremely short model update time for intelligent pumped storage power stations.
基金supported in part by grants from the National Institutes of Health (R01CA213149,R01CA241618).
文摘Nonlinear optical imaging is a versatile tool that has been proven to be exceptionally useful in various research fields.However,due to the use of photomultiplier tubes(PMTs),the wide application of nonlinear optical imaging is limited by the incapability of imaging under am-bient light.In this paper,we propose and demonstrate a new optical imaging detection method based on optical parametric amplification(OPA).As a nonlinear optical process,OPA in-trinsically rejects ambient light photons by coherence gating.Periodical poled lithium niobate(PPLN)crystals are used in this study as the media for OPA.Compared to bulk nonlinear optical crystals,PPLN crystals support the generation of OPA signal with lower pump power.Therefore,this characteristic of PPLN crystals is particularly beneficial when using high-repetition-rate lasers,which facilitate high-speed optical signal detection,such as in spec-troscopy and imaging.A PPLN-based OPA system was built to amplify the emitted imaging signal from second harmonic generation(SHG)and coherent anti-Stokes Raman scattering(CARS)microscopy imaging,and the amplified optical signal was strong enough to be detected by a biased photodiode under ordinary room light conditions.With OPA detection,ambient-light-on SHG and CARS imaging becomes possible,and achieves a similar result as PMT detection under strictly dark environments.These results demonstrate that OPA can be used as a substitute for PMTs in nonlinear optical imaging to adapt it to various applications with complex.light ing conditions.
文摘The Cloud system shows its growing functionalities in various industrial applications.The safety towards data transfer seems to be a threat where Network Intrusion Detection System(NIDS)is measured as an essential element to fulfill security.Recently,Machine Learning(ML)approaches have been used for the construction of intellectual IDS.Most IDS are based on ML techniques either as unsupervised or supervised.In supervised learning,NIDS is based on labeled data where it reduces the efficiency of the reduced model to identify attack patterns.Similarly,the unsupervised model fails to provide a satisfactory outcome.Hence,to boost the functionality of unsupervised learning,an effectual auto-encoder is applied for feature selection to select good features.Finally,the Naïve Bayes classifier is used for classification purposes.This approach exposes the finest generalization ability to train the data.The unlabelled data is also used for adoption towards data analysis.Here,redundant and noisy samples over the dataset are eliminated.To validate the robustness and efficiency of NIDS,the anticipated model is tested over the NSL-KDD dataset.The experimental outcomes demonstrate that the anticipated approach attains superior accuracy with 93%,which is higher compared to J48,AB tree,Random Forest(RF),Regression Tree(RT),Multi-Layer Perceptrons(MLP),Support Vector Machine(SVM),and Fuzzy.Similarly,False Alarm Rate(FAR)and True Positive Rate(TPR)of Naive Bayes(NB)is 0.3 and 0.99,respectively.When compared to prevailing techniques,the anticipated approach also delivers promising outcomes.