This paper aims to develop a nonrigid registration method of preoperative and intraoperative thoracoabdominal CT images in computer-assisted interventional surgeries for accurate tumor localization and tissue visualiz...This paper aims to develop a nonrigid registration method of preoperative and intraoperative thoracoabdominal CT images in computer-assisted interventional surgeries for accurate tumor localization and tissue visualization enhancement.However,fine structure registration of complex thoracoabdominal organs and large deformation registration caused by respiratory motion is challenging.To deal with this problem,we propose a 3D multi-scale attention VoxelMorph(MAVoxelMorph)registration network.To alleviate the large deformation problem,a multi-scale axial attention mechanism is utilized by using a residual dilated pyramid pooling for multi-scale feature extraction,and position-aware axial attention for long-distance dependencies between pixels capture.To further improve the large deformation and fine structure registration results,a multi-scale context channel attention mechanism is employed utilizing content information via adjacent encoding layers.Our method was evaluated on four public lung datasets(DIR-Lab dataset,Creatis dataset,Learn2Reg dataset,OASIS dataset)and a local dataset.Results proved that the proposed method achieved better registration performance than current state-of-the-art methods,especially in handling the registration of large deformations and fine structures.It also proved to be fast in 3D image registration,using about 1.5 s,and faster than most methods.Qualitative and quantitative assessments proved that the proposed MA-VoxelMorph has the potential to realize precise and fast tumor localization in clinical interventional surgeries.展开更多
COVID-19 posed challenges for global tourism management.Changes in visitor temporal and spatial patterns and their associated determinants pre-and peri-pandemic in Canadian Rocky Mountain National Parks are analyzed.D...COVID-19 posed challenges for global tourism management.Changes in visitor temporal and spatial patterns and their associated determinants pre-and peri-pandemic in Canadian Rocky Mountain National Parks are analyzed.Data was collected through social media programming and analyzed using spatiotemporal analysis and a geographically weighted regression(GWR)model.Results highlight that COVID-19 significantly changed park visitation patterns.Visitors tended to explore more remote areas peri-pandemic.The GWR model also indicated distance to nearby trails was a significant influence on visitor density.Our results indicate that the pandemic influenced tourism temporal and spatial imbalance.This research presents a novel approach using combined social media big data which can be extended to the field of tourism management,and has important implications to manage visitor patterns and to allocate resources efficiently to satisfy multiple objectives of park management.展开更多
The demand for Electronic Shelf Labels(ESL),according to the Internet of Things(IoT)paradigm,is expected to grow considerably in the immediate future.Various wireless communication standards are currently contending t...The demand for Electronic Shelf Labels(ESL),according to the Internet of Things(IoT)paradigm,is expected to grow considerably in the immediate future.Various wireless communication standards are currently contending to gain an edge over the competition and provide the massive connectivity that will be required by a world in which everyday objects are expected to communicate with each other.Low-Power Wide-Area Networks(LPWANs)are continuously gaining momentum among these standards,mainly thanks to their ability to provide long-range coverage to devices,exploiting license-free frequency bands.The main theme of this work is one of the most prominent LPWAN technologies,LoRa.The purpose of this research is to provide long-range,less intermediate node,less energy dissipation,and a cheaper ESL system.Much research has already been done on designing the LoRaWAN network,not capable to make a reliable network.LoRa is using different gateways to transmit the same data,collision,data jamming,and data repetition are expected.According to the transmission behavior of LoRa,50%of data is lost.In this paper,the Improved Backoff Algorithm with synchronization technique is used to decrease overlapping and data loss.Besides,the improved Adaptive Data Rate algorithm(ADR)avoids the collision in concurrently transmitted data by using different Spreading Factors(SFs).The allocation of SF has the main role in designing LoRa based network to minimize the impact of the intra-interference,cost function,and Euclidean distance.For this purpose,the K-means machine learning algorithm is used for clustering.The data rate model is using an intra-slicing technique based on Maximum Likelihood Estimation(MLE).The data rate model includes three critical communication slices,High Critical Communication(HCC),Medium Critical Communication(MCC),and Low Critical Communication(LCC),having the specified number of End devices(EDs),payload budget delay,and data rate.Finally,different combinations of gateways are used to build ESL for 200 electronic shelf labels.展开更多
A Recommender System(RS)is a crucial part of several firms,particularly those involved in e-commerce.In conventional RS,a user may only offer a single rating for an item-that is insufficient to perceive consumer prefe...A Recommender System(RS)is a crucial part of several firms,particularly those involved in e-commerce.In conventional RS,a user may only offer a single rating for an item-that is insufficient to perceive consumer preferences.Nowadays,businesses in industries like e-learning and tourism enable customers to rate a product using a variety of factors to comprehend customers’preferences.On the other hand,the collaborative filtering(CF)algorithm utilizing AutoEncoder(AE)is seen to be effective in identifying user-interested items.However,the cost of these computations increases nonlinearly as the number of items and users increases.To triumph over the issues,a novel expanded stacked autoencoder(ESAE)with Kernel Fuzzy C-Means Clustering(KFCM)technique is proposed with two phases.In the first phase of offline,the sparse multicriteria rating matrix is smoothened to a complete matrix by predicting the users’intact rating by the ESAE approach and users are clustered using the KFCM approach.In the next phase of online,the top-N recommendation prediction is made by the ESAE approach involving only the most similar user from multiple clusters.Hence the ESAE_KFCM model upgrades the prediction accuracy of 98.2%in Top-N recommendation with a minimized recommendation generation time.An experimental check on the Yahoo!Movies(YM)movie dataset and TripAdvisor(TA)travel dataset confirmed that the ESAE_KFCM model constantly outperforms conventional RS algorithms on a variety of assessment measures.展开更多
The security of the wireless sensor network-Internet of Things(WSN-IoT)network is more challenging due to its randomness and self-organized nature.Intrusion detection is one of the key methodologies utilized to ensure...The security of the wireless sensor network-Internet of Things(WSN-IoT)network is more challenging due to its randomness and self-organized nature.Intrusion detection is one of the key methodologies utilized to ensure the security of the network.Conventional intrusion detection mechanisms have issues such as higher misclassification rates,increased model complexity,insignificant feature extraction,increased training time,increased run time complexity,computation overhead,failure to identify new attacks,increased energy consumption,and a variety of other factors that limit the performance of the intrusion system model.In this research a security framework for WSN-IoT,through a deep learning technique is introduced using Modified Fuzzy-Adaptive DenseNet(MF_AdaDenseNet)and is benchmarked with datasets like NSL-KDD,UNSWNB15,CIDDS-001,Edge IIoT,Bot IoT.In this,the optimal feature selection using Capturing Dingo Optimization(CDO)is devised to acquire relevant features by removing redundant features.The proposed MF_AdaDenseNet intrusion detection model offers significant benefits by utilizing optimal feature selection with the CDO algorithm.This results in enhanced Detection Capacity with minimal computation complexity,as well as a reduction in False Alarm Rate(FAR)due to the consideration of classification error in the fitness estimation.As a result,the combined CDO-based feature selection and MF_AdaDenseNet intrusion detection mechanism outperform other state-of-the-art techniques,achieving maximal Detection Capacity,precision,recall,and F-Measure of 99.46%,99.54%,99.91%,and 99.68%,respectively,along with minimal FAR and Mean Absolute Error(MAE)of 0.9%and 0.11.展开更多
Achieving reliable and efficient weather classification for autonomous vehicles is crucial for ensuring safety and operational effectiveness.However,accurately classifying diverse and complex weather conditions remain...Achieving reliable and efficient weather classification for autonomous vehicles is crucial for ensuring safety and operational effectiveness.However,accurately classifying diverse and complex weather conditions remains a significant challenge.While advanced techniques such as Vision Transformers have been developed,they face key limitations,including high computational costs and limited generalization across varying weather conditions.These challenges present a critical research gap,particularly in applications where scalable and efficient solutions are needed to handle weather phenomena’intricate and dynamic nature in real-time.To address this gap,we propose a Multi-level Knowledge Distillation(MLKD)framework,which leverages the complementary strengths of state-of-the-art pre-trained models to enhance classification performance while minimizing computational overhead.Specifically,we employ ResNet50V2 and EfficientNetV2B3 as teacher models,known for their ability to capture complex image features and distil their knowledge into a custom lightweight Convolutional Neural Network(CNN)student model.This framework balances the trade-off between high classification accuracy and efficient resource consumption,ensuring real-time applicability in autonomous systems.Our Response-based Multi-level Knowledge Distillation(R-MLKD)approach effectively transfers rich,high-level feature representations from the teacher models to the student model,allowing the student to perform robustly with significantly fewer parameters and lower computational demands.The proposed method was evaluated on three public datasets(DAWN,BDD100K,and CITS traffic alerts),each containing seven weather classes with 2000 samples per class.The results demonstrate the effectiveness of MLKD,achieving a 97.3%accuracy,which surpasses conventional deep learning models.This work improves classification accuracy and tackles the practical challenges of model complexity,resource consumption,and real-time deployment,offering a scalable solution for weather classification in autonomous driving systems.展开更多
To satisfy the demand for good quality underwater welding and maintenance of nuclear power stations,a set of local dry automatic welding systems has been developed.These systems were based on an underwater robot that ...To satisfy the demand for good quality underwater welding and maintenance of nuclear power stations,a set of local dry automatic welding systems has been developed.These systems were based on an underwater robot that consisted of a special high-power underwater welding power supply,diving wire feeder,mini drain cap,welding robot,and special underwater welding torch.With a digital signal controller microprocessor as its core and combined with a dual inverter topology,the welding power supply was characterized by full-digital construction and multi-waveform flexible output.A compact diving wire feeding device was designed,based on the armature voltage negative feedback and high-frequency chopping pulse width modulation.This device yielded a high-efficiency seal of the driving motor with the help of dynamic and static sealing technology.To overcome the difficulty of local protection and deslagging in the welding area,a mini drain cap(with a duplexgas structure)based on the principle of the convergent nozzle was designed.The practical tests and the underwater welding experiments revealed that the underwater robotic local dry welding system is quite feasible.That is,the system could strike the arc stably and reliably in the shallow water environment,and formed beautiful welding seams.展开更多
Leukaemia is a type of blood cancer that is caused by undeveloped White Blood Cells(WBC),and it is also called a blast blood cell.In the marrow of human bones,leukaemia is developed and is responsible for blood cell g...Leukaemia is a type of blood cancer that is caused by undeveloped White Blood Cells(WBC),and it is also called a blast blood cell.In the marrow of human bones,leukaemia is developed and is responsible for blood cell generation with leukocytes and WBC,and if any cell gets blasted,then it may become a cause of death.Therefore,the diagnosis of leukaemia in its early stages helps greatly in the treatment along with saving human lives.Subsequently,in terms of detection,image segmentation techniques play a vital role,and they turn out to be the important image processing steps for the extraction of feature patterns from the Acute Lymphoblastic Leukaemia(ALL)type of blood cancer.Moreover,the image segmentation technique focuses on the division of cells by segmenting a microscopic image into background and cancer blood cell nucleus,which is well-known as the Region Of Interest(ROI).As a result,in this article,we attempt to build a segmentation technique capable of solving blood cell nucleus segmentation issues using four distinct scenarios,including K-means,FCM(Fuzzy Cmeans),K-means with FFA(Firefly Algorithm),and FCM with FFA.Also,we determine the most effective method of blood cell nucleus segmentation,which we subsequently use for the Leukaemia classification model.Finally,using the Convolution Neural Network(CNN)as a classifier,we developed a leukaemia cancer classification model from the microscopic images.The proposed system’s classification accuracy is tested using the CNN to test the model on the ALL-IDB dataset and equate it to the current state of the art.In terms of experimental analysis,we observed that the accuracy of the model is near to 99%,and it is far better than other existing models that are designed to segment and classify the types of leukaemia cancer in terms of ALL.展开更多
Medical image segmentation plays a crucial role in clinical diagnosis and therapy systems,yet still faces many challenges.Building on convolutional neural networks(CNNs),medical image segmentation has achieved tremend...Medical image segmentation plays a crucial role in clinical diagnosis and therapy systems,yet still faces many challenges.Building on convolutional neural networks(CNNs),medical image segmentation has achieved tremendous progress.However,owing to the locality of convolution operations,CNNs have the inherent limitation in learning global context.To address the limitation in building global context relationship from CNNs,we propose LGNet,a semantic segmentation network aiming to learn local and global features for fast and accurate medical image segmentation in this paper.Specifically,we employ a two-branch architecture consisting of convolution layers in one branch to learn local features and transformer layers in the other branch to learn global features.LGNet has two key insights:(1)We bridge two-branch to learn local and global features in an interactive way;(2)we present a novel multi-feature fusion model(MSFFM)to leverage the global contexture information from transformer and the local representational features from convolutions.Our method achieves state-of-the-art trade-off in terms of accuracy and efficiency on several medical image segmentation benchmarks including Synapse,ACDC and MOST.Specifically,LGNet achieves the state-of-the-art performance with Dice's indexes of 80.15%on Synapse,of 91.70%on ACDC,and of 95.56%on MOST.Meanwhile,the inference speed attains at 172 frames per second with 224-224 input resolution.The extensive experiments demonstrate the effectiveness of the proposed LGNet for fast and accurate for medical image segmentation.展开更多
Molecular motors are proteins or protein complexes which function as transporting engines in biological cells. This paper models the tether between motor and its cargo as a symmetric linear potential. Different from E...Molecular motors are proteins or protein complexes which function as transporting engines in biological cells. This paper models the tether between motor and its cargo as a symmetric linear potential. Different from Elston and Peskin＇s work for which performance of the system was discussed only in some limiting cases, this study produces analytic solutions of the problem for general cases by simplifying the transport system into two physical states, which makes it possible to discuss the dynamics of the motor--cargo system in detail. It turns out that the tether strength between motor and cargo should be greater than a threshold or the motor will fail to transport the cargo, which was not discussed by former researchers yet. Value of the threshold depends on the diffusion coefficients of cargo and motor and also on the strength of the Brownian ratchets dragging the system. The threshold approaches a finite constant when the strength of the ratchet tends to infinity.展开更多
The current advancement in cloud computing,Artificial Intelligence(AI),and the Internet of Things(IoT)transformed the traditional healthcare system into smart healthcare.Healthcare services could be enhanced by incorp...The current advancement in cloud computing,Artificial Intelligence(AI),and the Internet of Things(IoT)transformed the traditional healthcare system into smart healthcare.Healthcare services could be enhanced by incorporating key techniques like AI and IoT.The convergence of AI and IoT provides distinct opportunities in the medical field.Fall is regarded as a primary cause of death or post-traumatic complication for the ageing population.Therefore,earlier detection of older person falls in smart homes is required to improve the survival rate of an individual or provide the necessary support.Lately,the emergence of IoT,AI,smartphones,wearables,and so on making it possible to design fall detection(FD)systems for smart home care.This article introduces a new Teamwork Optimization with Deep Learning based Fall Detection for IoT Enabled Smart Healthcare Systems(TWODLFDSHS).The TWODL-FDSHS technique’s goal is to detect fall events for a smart healthcare system.Initially,the presented TWODL-FDSHS technique exploits IoT devices for the data collection process.Next,the TWODLFDSHS technique applies the TWO with Capsule Network(CapsNet)model for feature extraction.At last,a deep random vector functional link network(DRVFLN)with an Adam optimizer is exploited for fall event detection.A wide range of simulations took place to exhibit the enhanced performance of the presentedTWODL-FDSHS technique.The experimental outcomes stated the enhancements of the TWODL-FDSHS method over other models with increased accuracy of 98.30%on the URFD dataset.展开更多
White blood cells (WBC) or leukocytes are a vital component ofthe blood which forms the immune system, which is accountable to fightforeign elements. The WBC images can be exposed to different data analysisapproaches ...White blood cells (WBC) or leukocytes are a vital component ofthe blood which forms the immune system, which is accountable to fightforeign elements. The WBC images can be exposed to different data analysisapproaches which categorize different kinds of WBC. Conventionally, laboratorytests are carried out to determine the kind of WBC which is erroneousand time consuming. Recently, deep learning (DL) models can be employedfor automated investigation of WBC images in short duration. Therefore,this paper introduces an Aquila Optimizer with Transfer Learning basedAutomated White Blood Cells Classification (AOTL-WBCC) technique. Thepresented AOTL-WBCC model executes data normalization and data augmentationprocess (rotation and zooming) at the initial stage. In addition,the residual network (ResNet) approach was used for feature extraction inwhich the initial hyperparameter values of the ResNet model are tuned by theuse of AO algorithm. Finally, Bayesian neural network (BNN) classificationtechnique has been implied for the identification of WBC images into distinctclasses. The experimental validation of the AOTL-WBCC methodology isperformed with the help of Kaggle dataset. The experimental results foundthat the AOTL-WBCC model has outperformed other techniques which arebased on image processing and manual feature engineering approaches underdifferent dimensions.展开更多
Cloud computing utilizes enormous clusters of serviceable and manageable resources that can be virtually and dynamically recongured in order to deliver optimum resource utilization by exploiting the pay-per-use model....Cloud computing utilizes enormous clusters of serviceable and manageable resources that can be virtually and dynamically recongured in order to deliver optimum resource utilization by exploiting the pay-per-use model.However,concerns around security have been an impediment in the extensive adoption of the cloud computing model.In this regard,advancements in cryptography,accelerated by the wide usage of the internet worldwide,has emerged as a key area in addressing some of these security concerns.In this document,a hybrid cryptographic protocol deploying Blowsh and Paillier encryption algorithms has been presented and its strength compared with the existing hybrid Advanced Encryption Standard(AES)and Rivest Shamir Adleman(RSA)techniques.Algorithms for secure data storage protocol in two phases have been presented.The proposed hybrid protocol endeavors to improve the power of cloud storage through a decrease in computation time and ciphertext size.Simulations have been carried out with Oracle Virtual Box and Fog server used on an Ubuntu 16.04 platform.This grouping of asymmetric and homomorphic procedures has demonstrated enhanced security.Compression usage has helped in decreasing the storage space and computation time.Performance analysis in terms of computation overhead and quality of service parameters like loads of parameters with and without attacks,throughput,and stream length for different modes of block cipher mode has been carried out.Security analysis has been carried out by utilizing the Hardening Index as an audit parameter using Lynis 2.7.1.Similarly,for halting the aforementioned approaches and for regulating trafc,rewall protection has been generated in the chosen hybrid algorithms.Finally,enhancements in the performance of the Paillier and Blowsh hybrid scheme with and without compression compared to the existing schemes using RSA and AES procedures have been demonstrated.展开更多
Most of the existing security Mobicast routing protocols are not suitable for the monitoring applications with higher quality of service （QoS） requirement. A QoS dynamic clustering secure multicast scheme （QoS-DCSM...Most of the existing security Mobicast routing protocols are not suitable for the monitoring applications with higher quality of service （QoS） requirement. A QoS dynamic clustering secure multicast scheme （QoS-DCSMS） based on Mobicast and multi-level IxTESLA protocol for large-scale tracking sensornets is presented in this paper. The multicast clusters are dynamically formed according to the real-time status of nodes, and the cluster-head node is responsible for status review and certificating management of cluster nodes to ensure the most optimized QoS and security of multicast in this scheme. Another contribution of this paper is the optimal QoS security authentication algorithm, which analyzes the relationship between the QoS and the level Mofmulti-level oTESLA. Based on the analysis and simulation results, it shows that the influence to the network survival cycle （＇NSC） and real-time communication caused by energy consumption and latency in authentication is acceptable when the optimal QoS security authentication algorithm is satisfied.展开更多
Background A large number of robots have put forward the new requirements for human robot interaction.One of the problems in human-swarm robot interaction is how to naturally achieve an efficient and accurate interact...Background A large number of robots have put forward the new requirements for human robot interaction.One of the problems in human-swarm robot interaction is how to naturally achieve an efficient and accurate interaction between humans and swarm robot systems.To address this,this paper proposes a new type of human-swarm natural interaction system.Methods Through the cooperation between three-dimensional(3D)gesture interaction channel and natural language instruction channel,a natural and efficient interaction between a human and swarm robots is achieved.Results First,A 3D lasso technology realizes a batch-picking interaction of swarm robots through oriented bounding boxes.Second,control instruction labels for swarm-oriented robots are defined.The instruction label is integrated with the 3D gesture and natural language through instruction label filling.Finally,the understanding of natural language instructions is realized through a text classifier based on the maximum entropy model.A head-mounted augmented reality display device is used as a visual feedback channel.Conclusions The experiments on selecting robots verify the feasibility and availability of the system.展开更多
With cloud computing,large chunks of data can be handled at a small cost.However,there are some reservations regarding the security and privacy of cloud data stored.For solving these issues and enhancing cloud computi...With cloud computing,large chunks of data can be handled at a small cost.However,there are some reservations regarding the security and privacy of cloud data stored.For solving these issues and enhancing cloud computing security,this research provides a Three-Layered Security Access model(TLSA)aligned to an intrusion detection mechanism,access control mechanism,and data encryption system.The TLSA underlines the need for the protection of sensitive data.This proposed approach starts with Layer 1 data encryption using the Advanced Encryption Standard(AES).For data transfer and storage,this encryption guarantees the data’s authenticity and secrecy.Surprisingly,the solution employs the AES encryption algorithm to secure essential data before storing them in the Cloud to minimize unauthorized access.Role-based access control(RBAC)implements the second strategic level,which ensures specific personnel access certain data and resources.In RBAC,each user is allowed a specific role and Permission.This implies that permitted users can access some data stored in the Cloud.This layer assists in filtering granular access to data,reducing the risk that undesired data will be discovered during the process.Layer 3 deals with intrusion detection systems(IDS),which detect and quickly deal with malicious actions and intrusion attempts.The proposed TLSA security model of e-commerce includes conventional levels of security,such as encryption and access control,and encloses an insight intrusion detection system.This method offers integrated solutions for most typical security issues of cloud computing,including data secrecy,method of access,and threats.An extensive performance test was carried out to confirm the efficiency of the proposed three-tier security method.Comparisons have been made with state-of-art techniques,including DES,RSA,and DUAL-RSA,keeping into account Accuracy,QILV,F-Measure,Sensitivity,MSE,PSNR,SSIM,and computation time,encryption time,and decryption time.The proposed TLSA method provides an accuracy of 89.23%,F-Measure of 0.876,and SSIM of 0.8564 at a computation time of 5.7 s.A comparison with existing methods shows the better performance of the proposed method,thus confirming the enhanced ability to address security issues in cloud computing.展开更多
Time-triggered architecture,as a mainstream design of the distributed real-time system,has been successfully applied in the aerospace,automotive and mechanical industries.However,time-triggered scheduling is a challen...Time-triggered architecture,as a mainstream design of the distributed real-time system,has been successfully applied in the aerospace,automotive and mechanical industries.However,time-triggered scheduling is a challenging NP-hard problem.There are few studies that could quickly solve the scheduling problem of large distributed time-triggered systems.To solve this problem,a communication affinity parameter is defined in this paper to describe the degree of bias of the shaper task towards sending or receiving messages.Based on this,an innovative task-message decoupling model named D-scheduler is built to reduce the computation complexity of the scheduling problem in large-scale systems.Additionally,we provide mathematical proof that our model is a convex optimization that is easy to solve with existing computational tools.Our experiments substantiate the efficacy of the D-scheduler.It dramatically reduces the scheduling complexity of large-scale real-time systems with a small loss of solving space compared to the federal scheduler.展开更多
The advent of large scale models such as CLIP[1],GPT-2[2],and GPT-3[3]has marked a significant shift in the field of Artificial Intelligence(AI).Unlike the early days of Al,characterized by the design of distinct mode...The advent of large scale models such as CLIP[1],GPT-2[2],and GPT-3[3]has marked a significant shift in the field of Artificial Intelligence(AI).Unlike the early days of Al,characterized by the design of distinct models or the utilization of pre-trained models for fine-tuning in specific tasks,these modern models adopt a unified approach.展开更多
With the prevalence of Internet of Things(loT)devices,data collection has the potential to improve people's lives and create a significant value.However,it also exposes sensitive information,which leads to privacy...With the prevalence of Internet of Things(loT)devices,data collection has the potential to improve people's lives and create a significant value.However,it also exposes sensitive information,which leads to privacy risks.An approach called N-source anonymity has been used for privacy preservation in raw data collection,but most of the existing schemes do not have a balanced efficiency and robustness.In this work,a lightweight and efficient raw data collection scheme is proposed.The proposed scheme can not only collect data from the original users but also protect their privacy.Besides,the proposed scheme can resist user poisoning attacks,and the use of the reward method can motivate users to actively provide data.Analysis and simulation indicate that the proposed scheme is safe against poison attacks.Additionally,the proposed scheme has better performance in terms of computation and communication overhead compared to existing methods.High efficiency and appropriate incentive mechanisms indicate that the scheme is practical for IoT systems.展开更多
As one of the key technologies of intelligent vehicles, traffic sign detection is still a challenging task because of the tiny size of its target object. To address the challenge, we present a novel detection network ...As one of the key technologies of intelligent vehicles, traffic sign detection is still a challenging task because of the tiny size of its target object. To address the challenge, we present a novel detection network improved from yolo-v3 for the tiny traffic sign with high precision in real-time. First, a visual multi-scale attention module(MSAM), a light-weight yet effective module, is devised to fuse the multi-scale feature maps with channel weights and spatial masks. It increases the representation power of the network by emphasizing useful features and suppressing unnecessary ones. Second, we exploit effectively fine-grained features about tiny objects from the shallower layers through modifying backbone Darknet-53 and adding one prediction head to yolo-v3. Finally, a receptive field block is added into the neck of the network to broaden the receptive field. Experiments prove the effectiveness of our network in both quantitative and qualitative aspects. The m AP@0.5 of our network reaches 0.965 and its detection speed is55.56 FPS for 512 × 512 images on the challenging Tsinghua-Tencent 100 k(TT100 k) dataset.展开更多
基金supported in part by the National Natural Science Foundation of China[62301374]Hubei Provincial Natural Science Foundation of China[2022CFB804]+2 种基金Hubei Provincial Education Research Project[B2022057]the Youths Science Foundation of Wuhan Institute of Technology[K202240]the 15th Graduate Education Innovation Fund of Wuhan Institute of Technology[CX2023295].
文摘This paper aims to develop a nonrigid registration method of preoperative and intraoperative thoracoabdominal CT images in computer-assisted interventional surgeries for accurate tumor localization and tissue visualization enhancement.However,fine structure registration of complex thoracoabdominal organs and large deformation registration caused by respiratory motion is challenging.To deal with this problem,we propose a 3D multi-scale attention VoxelMorph(MAVoxelMorph)registration network.To alleviate the large deformation problem,a multi-scale axial attention mechanism is utilized by using a residual dilated pyramid pooling for multi-scale feature extraction,and position-aware axial attention for long-distance dependencies between pixels capture.To further improve the large deformation and fine structure registration results,a multi-scale context channel attention mechanism is employed utilizing content information via adjacent encoding layers.Our method was evaluated on four public lung datasets(DIR-Lab dataset,Creatis dataset,Learn2Reg dataset,OASIS dataset)and a local dataset.Results proved that the proposed method achieved better registration performance than current state-of-the-art methods,especially in handling the registration of large deformations and fine structures.It also proved to be fast in 3D image registration,using about 1.5 s,and faster than most methods.Qualitative and quantitative assessments proved that the proposed MA-VoxelMorph has the potential to realize precise and fast tumor localization in clinical interventional surgeries.
基金This research was supported by the UBC APFNet Grant(Project ID:2022sp2 CAN).
文摘COVID-19 posed challenges for global tourism management.Changes in visitor temporal and spatial patterns and their associated determinants pre-and peri-pandemic in Canadian Rocky Mountain National Parks are analyzed.Data was collected through social media programming and analyzed using spatiotemporal analysis and a geographically weighted regression(GWR)model.Results highlight that COVID-19 significantly changed park visitation patterns.Visitors tended to explore more remote areas peri-pandemic.The GWR model also indicated distance to nearby trails was a significant influence on visitor density.Our results indicate that the pandemic influenced tourism temporal and spatial imbalance.This research presents a novel approach using combined social media big data which can be extended to the field of tourism management,and has important implications to manage visitor patterns and to allocate resources efficiently to satisfy multiple objectives of park management.
基金This work is supported by the National Natural Science Foundation of China(61702020)Beijing Natural Science Foundation(4172013)Beijing Natural Science Foundation-Haidian Primitive Innovation Joint Fund(L182007).
文摘The demand for Electronic Shelf Labels(ESL),according to the Internet of Things(IoT)paradigm,is expected to grow considerably in the immediate future.Various wireless communication standards are currently contending to gain an edge over the competition and provide the massive connectivity that will be required by a world in which everyday objects are expected to communicate with each other.Low-Power Wide-Area Networks(LPWANs)are continuously gaining momentum among these standards,mainly thanks to their ability to provide long-range coverage to devices,exploiting license-free frequency bands.The main theme of this work is one of the most prominent LPWAN technologies,LoRa.The purpose of this research is to provide long-range,less intermediate node,less energy dissipation,and a cheaper ESL system.Much research has already been done on designing the LoRaWAN network,not capable to make a reliable network.LoRa is using different gateways to transmit the same data,collision,data jamming,and data repetition are expected.According to the transmission behavior of LoRa,50%of data is lost.In this paper,the Improved Backoff Algorithm with synchronization technique is used to decrease overlapping and data loss.Besides,the improved Adaptive Data Rate algorithm(ADR)avoids the collision in concurrently transmitted data by using different Spreading Factors(SFs).The allocation of SF has the main role in designing LoRa based network to minimize the impact of the intra-interference,cost function,and Euclidean distance.For this purpose,the K-means machine learning algorithm is used for clustering.The data rate model is using an intra-slicing technique based on Maximum Likelihood Estimation(MLE).The data rate model includes three critical communication slices,High Critical Communication(HCC),Medium Critical Communication(MCC),and Low Critical Communication(LCC),having the specified number of End devices(EDs),payload budget delay,and data rate.Finally,different combinations of gateways are used to build ESL for 200 electronic shelf labels.
文摘A Recommender System(RS)is a crucial part of several firms,particularly those involved in e-commerce.In conventional RS,a user may only offer a single rating for an item-that is insufficient to perceive consumer preferences.Nowadays,businesses in industries like e-learning and tourism enable customers to rate a product using a variety of factors to comprehend customers’preferences.On the other hand,the collaborative filtering(CF)algorithm utilizing AutoEncoder(AE)is seen to be effective in identifying user-interested items.However,the cost of these computations increases nonlinearly as the number of items and users increases.To triumph over the issues,a novel expanded stacked autoencoder(ESAE)with Kernel Fuzzy C-Means Clustering(KFCM)technique is proposed with two phases.In the first phase of offline,the sparse multicriteria rating matrix is smoothened to a complete matrix by predicting the users’intact rating by the ESAE approach and users are clustered using the KFCM approach.In the next phase of online,the top-N recommendation prediction is made by the ESAE approach involving only the most similar user from multiple clusters.Hence the ESAE_KFCM model upgrades the prediction accuracy of 98.2%in Top-N recommendation with a minimized recommendation generation time.An experimental check on the Yahoo!Movies(YM)movie dataset and TripAdvisor(TA)travel dataset confirmed that the ESAE_KFCM model constantly outperforms conventional RS algorithms on a variety of assessment measures.
基金Authors extend their appreciation to King Saud University for funding the publication of this research through the Researchers Supporting Project number(RSPD2024R809),King Saud University,Riyadh,Saudi Arabia.
文摘The security of the wireless sensor network-Internet of Things(WSN-IoT)network is more challenging due to its randomness and self-organized nature.Intrusion detection is one of the key methodologies utilized to ensure the security of the network.Conventional intrusion detection mechanisms have issues such as higher misclassification rates,increased model complexity,insignificant feature extraction,increased training time,increased run time complexity,computation overhead,failure to identify new attacks,increased energy consumption,and a variety of other factors that limit the performance of the intrusion system model.In this research a security framework for WSN-IoT,through a deep learning technique is introduced using Modified Fuzzy-Adaptive DenseNet(MF_AdaDenseNet)and is benchmarked with datasets like NSL-KDD,UNSWNB15,CIDDS-001,Edge IIoT,Bot IoT.In this,the optimal feature selection using Capturing Dingo Optimization(CDO)is devised to acquire relevant features by removing redundant features.The proposed MF_AdaDenseNet intrusion detection model offers significant benefits by utilizing optimal feature selection with the CDO algorithm.This results in enhanced Detection Capacity with minimal computation complexity,as well as a reduction in False Alarm Rate(FAR)due to the consideration of classification error in the fitness estimation.As a result,the combined CDO-based feature selection and MF_AdaDenseNet intrusion detection mechanism outperform other state-of-the-art techniques,achieving maximal Detection Capacity,precision,recall,and F-Measure of 99.46%,99.54%,99.91%,and 99.68%,respectively,along with minimal FAR and Mean Absolute Error(MAE)of 0.9%and 0.11.
文摘Achieving reliable and efficient weather classification for autonomous vehicles is crucial for ensuring safety and operational effectiveness.However,accurately classifying diverse and complex weather conditions remains a significant challenge.While advanced techniques such as Vision Transformers have been developed,they face key limitations,including high computational costs and limited generalization across varying weather conditions.These challenges present a critical research gap,particularly in applications where scalable and efficient solutions are needed to handle weather phenomena’intricate and dynamic nature in real-time.To address this gap,we propose a Multi-level Knowledge Distillation(MLKD)framework,which leverages the complementary strengths of state-of-the-art pre-trained models to enhance classification performance while minimizing computational overhead.Specifically,we employ ResNet50V2 and EfficientNetV2B3 as teacher models,known for their ability to capture complex image features and distil their knowledge into a custom lightweight Convolutional Neural Network(CNN)student model.This framework balances the trade-off between high classification accuracy and efficient resource consumption,ensuring real-time applicability in autonomous systems.Our Response-based Multi-level Knowledge Distillation(R-MLKD)approach effectively transfers rich,high-level feature representations from the teacher models to the student model,allowing the student to perform robustly with significantly fewer parameters and lower computational demands.The proposed method was evaluated on three public datasets(DAWN,BDD100K,and CITS traffic alerts),each containing seven weather classes with 2000 samples per class.The results demonstrate the effectiveness of MLKD,achieving a 97.3%accuracy,which surpasses conventional deep learning models.This work improves classification accuracy and tackles the practical challenges of model complexity,resource consumption,and real-time deployment,offering a scalable solution for weather classification in autonomous driving systems.
基金This work was supported by the National Natural Science Foundation of China(Grant Numbers 51375173).
文摘To satisfy the demand for good quality underwater welding and maintenance of nuclear power stations,a set of local dry automatic welding systems has been developed.These systems were based on an underwater robot that consisted of a special high-power underwater welding power supply,diving wire feeder,mini drain cap,welding robot,and special underwater welding torch.With a digital signal controller microprocessor as its core and combined with a dual inverter topology,the welding power supply was characterized by full-digital construction and multi-waveform flexible output.A compact diving wire feeding device was designed,based on the armature voltage negative feedback and high-frequency chopping pulse width modulation.This device yielded a high-efficiency seal of the driving motor with the help of dynamic and static sealing technology.To overcome the difficulty of local protection and deslagging in the welding area,a mini drain cap(with a duplexgas structure)based on the principle of the convergent nozzle was designed.The practical tests and the underwater welding experiments revealed that the underwater robotic local dry welding system is quite feasible.That is,the system could strike the arc stably and reliably in the shallow water environment,and formed beautiful welding seams.
基金We deeply acknowledge Taif University for supporting this study through Taif University Researchers Supporting Project number(TURSP-2020/115),Taif University,Taif,Saudi Arabia.
文摘Leukaemia is a type of blood cancer that is caused by undeveloped White Blood Cells(WBC),and it is also called a blast blood cell.In the marrow of human bones,leukaemia is developed and is responsible for blood cell generation with leukocytes and WBC,and if any cell gets blasted,then it may become a cause of death.Therefore,the diagnosis of leukaemia in its early stages helps greatly in the treatment along with saving human lives.Subsequently,in terms of detection,image segmentation techniques play a vital role,and they turn out to be the important image processing steps for the extraction of feature patterns from the Acute Lymphoblastic Leukaemia(ALL)type of blood cancer.Moreover,the image segmentation technique focuses on the division of cells by segmenting a microscopic image into background and cancer blood cell nucleus,which is well-known as the Region Of Interest(ROI).As a result,in this article,we attempt to build a segmentation technique capable of solving blood cell nucleus segmentation issues using four distinct scenarios,including K-means,FCM(Fuzzy Cmeans),K-means with FFA(Firefly Algorithm),and FCM with FFA.Also,we determine the most effective method of blood cell nucleus segmentation,which we subsequently use for the Leukaemia classification model.Finally,using the Convolution Neural Network(CNN)as a classifier,we developed a leukaemia cancer classification model from the microscopic images.The proposed system’s classification accuracy is tested using the CNN to test the model on the ALL-IDB dataset and equate it to the current state of the art.In terms of experimental analysis,we observed that the accuracy of the model is near to 99%,and it is far better than other existing models that are designed to segment and classify the types of leukaemia cancer in terms of ALL.
基金supported by the Open-Fund of WNLO (Grant No.2018WNLOKF027)the Hubei Key Laboratory of Intelligent Robot in Wuhan Institute of Technology (Grant No.HBIRL 202003).
文摘Medical image segmentation plays a crucial role in clinical diagnosis and therapy systems,yet still faces many challenges.Building on convolutional neural networks(CNNs),medical image segmentation has achieved tremendous progress.However,owing to the locality of convolution operations,CNNs have the inherent limitation in learning global context.To address the limitation in building global context relationship from CNNs,we propose LGNet,a semantic segmentation network aiming to learn local and global features for fast and accurate medical image segmentation in this paper.Specifically,we employ a two-branch architecture consisting of convolution layers in one branch to learn local features and transformer layers in the other branch to learn global features.LGNet has two key insights:(1)We bridge two-branch to learn local and global features in an interactive way;(2)we present a novel multi-feature fusion model(MSFFM)to leverage the global contexture information from transformer and the local representational features from convolutions.Our method achieves state-of-the-art trade-off in terms of accuracy and efficiency on several medical image segmentation benchmarks including Synapse,ACDC and MOST.Specifically,LGNet achieves the state-of-the-art performance with Dice's indexes of 80.15%on Synapse,of 91.70%on ACDC,and of 95.56%on MOST.Meanwhile,the inference speed attains at 172 frames per second with 224-224 input resolution.The extensive experiments demonstrate the effectiveness of the proposed LGNet for fast and accurate for medical image segmentation.
基金supported by the National Natural Science Foundation of China (Grant No. 30600121)Doctoral Foundation of Shandong Province of China (Grant No. 2007BS09002)
文摘Molecular motors are proteins or protein complexes which function as transporting engines in biological cells. This paper models the tether between motor and its cargo as a symmetric linear potential. Different from Elston and Peskin＇s work for which performance of the system was discussed only in some limiting cases, this study produces analytic solutions of the problem for general cases by simplifying the transport system into two physical states, which makes it possible to discuss the dynamics of the motor--cargo system in detail. It turns out that the tether strength between motor and cargo should be greater than a threshold or the motor will fail to transport the cargo, which was not discussed by former researchers yet. Value of the threshold depends on the diffusion coefficients of cargo and motor and also on the strength of the Brownian ratchets dragging the system. The threshold approaches a finite constant when the strength of the ratchet tends to infinity.
基金The Deanship of Scientific Research (DSR)at King Abdulaziz University (KAU),Jeddah,Saudi Arabia has funded this project,under grant no.KEP-4-120-42.
文摘The current advancement in cloud computing,Artificial Intelligence(AI),and the Internet of Things(IoT)transformed the traditional healthcare system into smart healthcare.Healthcare services could be enhanced by incorporating key techniques like AI and IoT.The convergence of AI and IoT provides distinct opportunities in the medical field.Fall is regarded as a primary cause of death or post-traumatic complication for the ageing population.Therefore,earlier detection of older person falls in smart homes is required to improve the survival rate of an individual or provide the necessary support.Lately,the emergence of IoT,AI,smartphones,wearables,and so on making it possible to design fall detection(FD)systems for smart home care.This article introduces a new Teamwork Optimization with Deep Learning based Fall Detection for IoT Enabled Smart Healthcare Systems(TWODLFDSHS).The TWODL-FDSHS technique’s goal is to detect fall events for a smart healthcare system.Initially,the presented TWODL-FDSHS technique exploits IoT devices for the data collection process.Next,the TWODLFDSHS technique applies the TWO with Capsule Network(CapsNet)model for feature extraction.At last,a deep random vector functional link network(DRVFLN)with an Adam optimizer is exploited for fall event detection.A wide range of simulations took place to exhibit the enhanced performance of the presentedTWODL-FDSHS technique.The experimental outcomes stated the enhancements of the TWODL-FDSHS method over other models with increased accuracy of 98.30%on the URFD dataset.
基金The Deanship of Scientific Research (DSR)at King Abdulaziz University (KAU),Jeddah,Saudi Arabia has funded this project,under Grant No.KEP-1–120–42.
文摘White blood cells (WBC) or leukocytes are a vital component ofthe blood which forms the immune system, which is accountable to fightforeign elements. The WBC images can be exposed to different data analysisapproaches which categorize different kinds of WBC. Conventionally, laboratorytests are carried out to determine the kind of WBC which is erroneousand time consuming. Recently, deep learning (DL) models can be employedfor automated investigation of WBC images in short duration. Therefore,this paper introduces an Aquila Optimizer with Transfer Learning basedAutomated White Blood Cells Classification (AOTL-WBCC) technique. Thepresented AOTL-WBCC model executes data normalization and data augmentationprocess (rotation and zooming) at the initial stage. In addition,the residual network (ResNet) approach was used for feature extraction inwhich the initial hyperparameter values of the ResNet model are tuned by theuse of AO algorithm. Finally, Bayesian neural network (BNN) classificationtechnique has been implied for the identification of WBC images into distinctclasses. The experimental validation of the AOTL-WBCC methodology isperformed with the help of Kaggle dataset. The experimental results foundthat the AOTL-WBCC model has outperformed other techniques which arebased on image processing and manual feature engineering approaches underdifferent dimensions.
文摘Cloud computing utilizes enormous clusters of serviceable and manageable resources that can be virtually and dynamically recongured in order to deliver optimum resource utilization by exploiting the pay-per-use model.However,concerns around security have been an impediment in the extensive adoption of the cloud computing model.In this regard,advancements in cryptography,accelerated by the wide usage of the internet worldwide,has emerged as a key area in addressing some of these security concerns.In this document,a hybrid cryptographic protocol deploying Blowsh and Paillier encryption algorithms has been presented and its strength compared with the existing hybrid Advanced Encryption Standard(AES)and Rivest Shamir Adleman(RSA)techniques.Algorithms for secure data storage protocol in two phases have been presented.The proposed hybrid protocol endeavors to improve the power of cloud storage through a decrease in computation time and ciphertext size.Simulations have been carried out with Oracle Virtual Box and Fog server used on an Ubuntu 16.04 platform.This grouping of asymmetric and homomorphic procedures has demonstrated enhanced security.Compression usage has helped in decreasing the storage space and computation time.Performance analysis in terms of computation overhead and quality of service parameters like loads of parameters with and without attacks,throughput,and stream length for different modes of block cipher mode has been carried out.Security analysis has been carried out by utilizing the Hardening Index as an audit parameter using Lynis 2.7.1.Similarly,for halting the aforementioned approaches and for regulating trafc,rewall protection has been generated in the chosen hybrid algorithms.Finally,enhancements in the performance of the Paillier and Blowsh hybrid scheme with and without compression compared to the existing schemes using RSA and AES procedures have been demonstrated.
基金Supported by the National Natural Science Foundation of China （No. 60903157）
文摘Most of the existing security Mobicast routing protocols are not suitable for the monitoring applications with higher quality of service （QoS） requirement. A QoS dynamic clustering secure multicast scheme （QoS-DCSMS） based on Mobicast and multi-level IxTESLA protocol for large-scale tracking sensornets is presented in this paper. The multicast clusters are dynamically formed according to the real-time status of nodes, and the cluster-head node is responsible for status review and certificating management of cluster nodes to ensure the most optimized QoS and security of multicast in this scheme. Another contribution of this paper is the optimal QoS security authentication algorithm, which analyzes the relationship between the QoS and the level Mofmulti-level oTESLA. Based on the analysis and simulation results, it shows that the influence to the network survival cycle （＇NSC） and real-time communication caused by energy consumption and latency in authentication is acceptable when the optimal QoS security authentication algorithm is satisfied.
基金Key-Area Research and Development Program of Guangdong Province(2019B090915002).
文摘Background A large number of robots have put forward the new requirements for human robot interaction.One of the problems in human-swarm robot interaction is how to naturally achieve an efficient and accurate interaction between humans and swarm robot systems.To address this,this paper proposes a new type of human-swarm natural interaction system.Methods Through the cooperation between three-dimensional(3D)gesture interaction channel and natural language instruction channel,a natural and efficient interaction between a human and swarm robots is achieved.Results First,A 3D lasso technology realizes a batch-picking interaction of swarm robots through oriented bounding boxes.Second,control instruction labels for swarm-oriented robots are defined.The instruction label is integrated with the 3D gesture and natural language through instruction label filling.Finally,the understanding of natural language instructions is realized through a text classifier based on the maximum entropy model.A head-mounted augmented reality display device is used as a visual feedback channel.Conclusions The experiments on selecting robots verify the feasibility and availability of the system.
基金funded by UKRI EPSRC Grant EP/W020408/1 Project SPRITE+2:The Security,Privacy,Identity and Trust Engagement Network plus(phase 2)for this studyThe authors also have been funded by PhD project RS718 on Explainable AI through UKRI EPSRC Grant funded Doctoral Training Centre at Swansea University.
文摘With cloud computing,large chunks of data can be handled at a small cost.However,there are some reservations regarding the security and privacy of cloud data stored.For solving these issues and enhancing cloud computing security,this research provides a Three-Layered Security Access model(TLSA)aligned to an intrusion detection mechanism,access control mechanism,and data encryption system.The TLSA underlines the need for the protection of sensitive data.This proposed approach starts with Layer 1 data encryption using the Advanced Encryption Standard(AES).For data transfer and storage,this encryption guarantees the data’s authenticity and secrecy.Surprisingly,the solution employs the AES encryption algorithm to secure essential data before storing them in the Cloud to minimize unauthorized access.Role-based access control(RBAC)implements the second strategic level,which ensures specific personnel access certain data and resources.In RBAC,each user is allowed a specific role and Permission.This implies that permitted users can access some data stored in the Cloud.This layer assists in filtering granular access to data,reducing the risk that undesired data will be discovered during the process.Layer 3 deals with intrusion detection systems(IDS),which detect and quickly deal with malicious actions and intrusion attempts.The proposed TLSA security model of e-commerce includes conventional levels of security,such as encryption and access control,and encloses an insight intrusion detection system.This method offers integrated solutions for most typical security issues of cloud computing,including data secrecy,method of access,and threats.An extensive performance test was carried out to confirm the efficiency of the proposed three-tier security method.Comparisons have been made with state-of-art techniques,including DES,RSA,and DUAL-RSA,keeping into account Accuracy,QILV,F-Measure,Sensitivity,MSE,PSNR,SSIM,and computation time,encryption time,and decryption time.The proposed TLSA method provides an accuracy of 89.23%,F-Measure of 0.876,and SSIM of 0.8564 at a computation time of 5.7 s.A comparison with existing methods shows the better performance of the proposed method,thus confirming the enhanced ability to address security issues in cloud computing.
基金partially supported by the National Natural Science Foundation of China(Grant Nos.62176016 and 72274127)the National Key R&D Program of China(Grant No.2021YFB2104800)+3 种基金Guizhou Province Science and Technology Project:Research and Demonstration of Sci.Tech Big Data Mining Technology Based on Knowledge Graph(supported by Qiankehe[2021]General 382)Teaching Reform Project of Beihang University in 2020:Standardized Teaching and Intelligent Analysis System Construction for Production PracticeCapital Health Development Research Project(Grant No.2022-2-2013)the Young Talent Development Grant of Beijing Economic-Technological Development Area(Grant No.2140030001870)。
文摘Time-triggered architecture,as a mainstream design of the distributed real-time system,has been successfully applied in the aerospace,automotive and mechanical industries.However,time-triggered scheduling is a challenging NP-hard problem.There are few studies that could quickly solve the scheduling problem of large distributed time-triggered systems.To solve this problem,a communication affinity parameter is defined in this paper to describe the degree of bias of the shaper task towards sending or receiving messages.Based on this,an innovative task-message decoupling model named D-scheduler is built to reduce the computation complexity of the scheduling problem in large-scale systems.Additionally,we provide mathematical proof that our model is a convex optimization that is easy to solve with existing computational tools.Our experiments substantiate the efficacy of the D-scheduler.It dramatically reduces the scheduling complexity of large-scale real-time systems with a small loss of solving space compared to the federal scheduler.
基金This work was supported by the National Natural Science Foundation of China(Grant No.62206048)Natural Science Foundation of Jiangsu Province(BK20220819)+1 种基金Young Elite Scientists Sponsorship Program of Jiangsu Association for Science and Technology(Tj-2022-027)the Big Data Computing Center of Southeast University.
文摘The advent of large scale models such as CLIP[1],GPT-2[2],and GPT-3[3]has marked a significant shift in the field of Artificial Intelligence(AI).Unlike the early days of Al,characterized by the design of distinct models or the utilization of pre-trained models for fine-tuning in specific tasks,these modern models adopt a unified approach.
基金supported in part by the National Natural Science Foundation of China(62072133)the Innovation Project of Guangxi Graduate Education(YCSW2022279)Wenzhou Science and Technology Plan(2023ZW0013).
文摘With the prevalence of Internet of Things(loT)devices,data collection has the potential to improve people's lives and create a significant value.However,it also exposes sensitive information,which leads to privacy risks.An approach called N-source anonymity has been used for privacy preservation in raw data collection,but most of the existing schemes do not have a balanced efficiency and robustness.In this work,a lightweight and efficient raw data collection scheme is proposed.The proposed scheme can not only collect data from the original users but also protect their privacy.Besides,the proposed scheme can resist user poisoning attacks,and the use of the reward method can motivate users to actively provide data.Analysis and simulation indicate that the proposed scheme is safe against poison attacks.Additionally,the proposed scheme has better performance in terms of computation and communication overhead compared to existing methods.High efficiency and appropriate incentive mechanisms indicate that the scheme is practical for IoT systems.
基金supported by the National Key R&D Program of China(Grant Nos.2018YFB2101100 and 2019YFB2101600)the National Natural Science Foundation of China(Grant No.62176016)+2 种基金the Guizhou Province Science and Technology Project:Research and Demonstration of Science and Technology Big Data Mining Technology Based on Knowledge Graph(Qiankehe[2021]General 382)the Training Program of the Major Research Plan of the National Natural Science Foundation of China(Grant No.92046015)the Beijing Natural Science Foundation Program and Scientific Research Key Program of Beijing Municipal Commission of Education(Grant No.KZ202010025047)。
文摘As one of the key technologies of intelligent vehicles, traffic sign detection is still a challenging task because of the tiny size of its target object. To address the challenge, we present a novel detection network improved from yolo-v3 for the tiny traffic sign with high precision in real-time. First, a visual multi-scale attention module(MSAM), a light-weight yet effective module, is devised to fuse the multi-scale feature maps with channel weights and spatial masks. It increases the representation power of the network by emphasizing useful features and suppressing unnecessary ones. Second, we exploit effectively fine-grained features about tiny objects from the shallower layers through modifying backbone Darknet-53 and adding one prediction head to yolo-v3. Finally, a receptive field block is added into the neck of the network to broaden the receptive field. Experiments prove the effectiveness of our network in both quantitative and qualitative aspects. The m AP@0.5 of our network reaches 0.965 and its detection speed is55.56 FPS for 512 × 512 images on the challenging Tsinghua-Tencent 100 k(TT100 k) dataset.
