Human pose estimation aims to localize the body joints from image or video data.With the development of deeplearning,pose estimation has become a hot research topic in the field of computer vision.In recent years,huma...Human pose estimation aims to localize the body joints from image or video data.With the development of deeplearning,pose estimation has become a hot research topic in the field of computer vision.In recent years,humanpose estimation has achieved great success in multiple fields such as animation and sports.However,to obtainaccurate positioning results,existing methods may suffer from large model sizes,a high number of parameters,and increased complexity,leading to high computing costs.In this paper,we propose a new lightweight featureencoder to construct a high-resolution network that reduces the number of parameters and lowers the computingcost.We also introduced a semantic enhancement module that improves global feature extraction and networkperformance by combining channel and spatial dimensions.Furthermore,we propose a dense connected spatialpyramid pooling module to compensate for the decrease in image resolution and information loss in the network.Finally,ourmethod effectively reduces the number of parameters and complexitywhile ensuring high performance.Extensive experiments show that our method achieves a competitive performance while dramatically reducing thenumber of parameters,and operational complexity.Specifically,our method can obtain 89.9%AP score on MPIIVAL,while the number of parameters and the complexity of operations were reduced by 41%and 36%,respectively.展开更多
With the advancement of wireless network technology,vast amounts of traffic have been generated,and malicious traffic attacks that threaten the network environment are becoming increasingly sophisticated.While signatu...With the advancement of wireless network technology,vast amounts of traffic have been generated,and malicious traffic attacks that threaten the network environment are becoming increasingly sophisticated.While signature-based detection methods,static analysis,and dynamic analysis techniques have been previously explored for malicious traffic detection,they have limitations in identifying diversified malware traffic patterns.Recent research has been focused on the application of machine learning to detect these patterns.However,applying machine learning to lightweight devices like IoT devices is challenging because of the high computational demands and complexity involved in the learning process.In this study,we examined methods for effectively utilizing machine learning-based malicious traffic detection approaches for lightweight devices.We introduced the suboptimal feature selection model(SFSM),a feature selection technique designed to reduce complexity while maintaining the effectiveness of malicious traffic detection.Detection performance was evaluated on various malicious traffic,benign,exploits,and generic,using the UNSW-NB15 dataset and SFSM sub-optimized hyperparameters for feature selection and narrowed the search scope to encompass all features.SFSM improved learning performance while minimizing complexity by considering feature selection and exhaustive search as two steps,a problem not considered in conventional models.Our experimental results showed that the detection accuracy was improved by approximately 20%compared to the random model,and the reduction in accuracy compared to the greedy model,which performs an exhaustive search on all features,was kept within 6%.Additionally,latency and complexity were reduced by approximately 96%and 99.78%,respectively,compared to the greedy model.This study demonstrates that malicious traffic can be effectively detected even in lightweight device environments.SFSM verified the possibility of detecting various attack traffic on lightweight devices.展开更多
Galloping cheetahs,climbing mountain goats,and load hauling horses all show desirable locomotion capability,which motivates the development of quadruped robots.Among various quadruped robots,hydraulically driven quadr...Galloping cheetahs,climbing mountain goats,and load hauling horses all show desirable locomotion capability,which motivates the development of quadruped robots.Among various quadruped robots,hydraulically driven quadruped robots show great potential in unstructured environments due to their discrete landing positions and large payloads.As the most critical movement unit of a quadruped robot,the limb leg unit(LLU)directly affects movement speed and reliability,and requires a compact and lightweight design.Inspired by the dexterous skeleton–muscle systems of cheetahs and humans,this paper proposes a highly integrated bionic actuator system for a better dynamic performance of an LLU.We propose that a cylinder barrel with multiple element interfaces and internal smooth channels is realized using metal additive manufacturing,and hybrid lattice structures are introduced into the lightweight design of the piston rod.In addition,additive manufacturing and topology optimization are incorporated to reduce the redundant material of the structural parts of the LLU.The mechanical properties of the actuator system are verified by numerical simulation and experiments,and the power density of the actuators is far greater than that of cheetah muscle.The mass of the optimized LLU is reduced by 24.5%,and the optimized LLU shows better response time performance when given a step signal,and presents a good trajectory tracking ability with the increase in motion frequency.展开更多
The widespread and growing interest in the Internet of Things(IoT)may be attributed to its usefulness in many different fields.Physical settings are probed for data,which is then transferred via linked networks.There ...The widespread and growing interest in the Internet of Things(IoT)may be attributed to its usefulness in many different fields.Physical settings are probed for data,which is then transferred via linked networks.There are several hurdles to overcome when putting IoT into practice,from managing server infrastructure to coordinating the use of tiny sensors.When it comes to deploying IoT,everyone agrees that security is the biggest issue.This is due to the fact that a large number of IoT devices exist in the physicalworld and thatmany of themhave constrained resources such as electricity,memory,processing power,and square footage.This research intends to analyse resource-constrained IoT devices,including RFID tags,sensors,and smart cards,and the issues involved with protecting them in such restricted circumstances.Using lightweight cryptography,the information sent between these gadgets may be secured.In order to provide a holistic picture,this research evaluates and contrasts well-known algorithms based on their implementation cost,hardware/software efficiency,and attack resistance features.We also emphasised how essential lightweight encryption is for striking a good cost-to-performance-to-security ratio.展开更多
Automatic crack detection of cement pavement chiefly benefits from the rapid development of deep learning,with convolutional neural networks(CNN)playing an important role in this field.However,as the performance of cr...Automatic crack detection of cement pavement chiefly benefits from the rapid development of deep learning,with convolutional neural networks(CNN)playing an important role in this field.However,as the performance of crack detection in cement pavement improves,the depth and width of the network structure are significantly increased,which necessitates more computing power and storage space.This limitation hampers the practical implementation of crack detection models on various platforms,particularly portable devices like small mobile devices.To solve these problems,we propose a dual-encoder-based network architecture that focuses on extracting more comprehensive fracture feature information and combines cross-fusion modules and coordinated attention mechanisms formore efficient feature fusion.Firstly,we use small channel convolution to construct shallow feature extractionmodule(SFEM)to extract low-level feature information of cracks in cement pavement images,in order to obtainmore information about cracks in the shallowfeatures of images.In addition,we construct large kernel atrous convolution(LKAC)to enhance crack information,which incorporates coordination attention mechanism for non-crack information filtering,and large kernel atrous convolution with different cores,using different receptive fields to extract more detailed edge and context information.Finally,the three-stage feature map outputs from the shallow feature extraction module is cross-fused with the two-stage feature map outputs from the large kernel atrous convolution module,and the shallow feature and detailed edge feature are fully fused to obtain the final crack prediction map.We evaluate our method on three public crack datasets:DeepCrack,CFD,and Crack500.Experimental results on theDeepCrack dataset demonstrate the effectiveness of our proposed method compared to state-of-the-art crack detection methods,which achieves Precision(P)87.2%,Recall(R)87.7%,and F-score(F1)87.4%.Thanks to our lightweight crack detectionmodel,the parameter count of the model in real-world detection scenarios has been significantly reduced to less than 2M.This advancement also facilitates technical support for portable scene detection.展开更多
Blockchain technology has garnered significant attention from global organizations and researchers due to its potential as a solution for centralized system challenges.Concurrently,the Internet of Things(IoT)has revol...Blockchain technology has garnered significant attention from global organizations and researchers due to its potential as a solution for centralized system challenges.Concurrently,the Internet of Things(IoT)has revolutionized the Fourth Industrial Revolution by enabling interconnected devices to offer innovative services,ultimately enhancing human lives.This paper presents a new approach utilizing lightweight blockchain technology,effectively reducing the computational burden typically associated with conventional blockchain systems.By integrating this lightweight blockchain with IoT systems,substantial reductions in implementation time and computational complexity can be achieved.Moreover,the paper proposes the utilization of the Okamoto Uchiyama encryption algorithm,renowned for its homomorphic characteristics,to reinforce the privacy and security of IoT-generated data.The integration of homomorphic encryption and blockchain technology establishes a secure and decentralized platformfor storing and analyzing sensitive data of the supply chain data.This platformfacilitates the development of some business models and empowers decentralized applications to perform computations on encrypted data while maintaining data privacy.The results validate the robust security of the proposed system,comparable to standard blockchain implementations,leveraging the distinctive homomorphic attributes of the Okamoto Uchiyama algorithm and the lightweight blockchain paradigm.展开更多
With the growth of the Internet,more and more business is being done online,for example,online offices,online education and so on.While this makes people’s lives more convenient,it also increases the risk of the netw...With the growth of the Internet,more and more business is being done online,for example,online offices,online education and so on.While this makes people’s lives more convenient,it also increases the risk of the network being attacked by malicious code.Therefore,it is important to identify malicious codes on computer systems efficiently.However,most of the existing malicious code detection methods have two problems:(1)The ability of the model to extract features is weak,resulting in poor model performance.(2)The large scale of model data leads to difficulties deploying on devices with limited resources.Therefore,this paper proposes a lightweight malicious code identification model Lightweight Malicious Code Classification Method Based on Improved SqueezeNet(LCMISNet).In this paper,the MFire lightweight feature extraction module is constructed by proposing a feature slicing module and a multi-size depthwise separable convolution module.The feature slicing module reduces the number of parameters by grouping features.The multi-size depthwise separable convolution module reduces the number of parameters and enhances the feature extraction capability by replacing the standard convolution with depthwise separable convolution with different convolution kernel sizes.In addition,this paper also proposes a feature splicing module to connect the MFire lightweight feature extraction module based on the feature reuse and constructs the lightweight model LCMISNet.The malicious code recognition accuracy of LCMISNet on the BIG 2015 dataset and the Malimg dataset reaches 98.90% and 99.58%,respectively.It proves that LCMISNet has a powerful malicious code recognition performance.In addition,compared with other network models,LCMISNet has better performance,and a lower number of parameters and computations.展开更多
A lightweight malware detection and family classification system for the Internet of Things (IoT) was designed to solve the difficulty of deploying defense models caused by the limited computing and storage resources ...A lightweight malware detection and family classification system for the Internet of Things (IoT) was designed to solve the difficulty of deploying defense models caused by the limited computing and storage resources of IoT devices. By training complex models with IoT software gray-scale images and utilizing the gradient-weighted class-activated mapping technique, the system can identify key codes that influence model decisions. This allows for the reconstruction of gray-scale images to train a lightweight model called LMDNet for malware detection. Additionally, the multi-teacher knowledge distillation method is employed to train KD-LMDNet, which focuses on classifying malware families. The results indicate that the model’s identification speed surpasses that of traditional methods by 23.68%. Moreover, the accuracy achieved on the Malimg dataset for family classification is an impressive 99.07%. Furthermore, with a model size of only 0.45M, it appears to be well-suited for the IoT environment. By training complex models using IoT software gray-scale images and utilizing the gradient-weighted class-activated mapping technique, the system can identify key codes that influence model decisions. This allows for the reconstruction of gray-scale images to train a lightweight model called LMDNet for malware detection. Thus, the presented approach can address the challenges associated with malware detection and family classification in IoT devices.展开更多
Convolutional neural networks (CNNs) are widely used in image classification tasks, but their increasing model size and computation make them challenging to implement on embedded systems with constrained hardware reso...Convolutional neural networks (CNNs) are widely used in image classification tasks, but their increasing model size and computation make them challenging to implement on embedded systems with constrained hardware resources. To address this issue, the MobileNetV1 network was developed, which employs depthwise convolution to reduce network complexity. MobileNetV1 employs a stride of 2 in several convolutional layers to decrease the spatial resolution of feature maps, thereby lowering computational costs. However, this stride setting can lead to a loss of spatial information, particularly affecting the detection and representation of smaller objects or finer details in images. To maintain the trade-off between complexity and model performance, a lightweight convolutional neural network with hierarchical multi-scale feature fusion based on the MobileNetV1 network is proposed. The network consists of two main subnetworks. The first subnetwork uses a depthwise dilated separable convolution (DDSC) layer to learn imaging features with fewer parameters, which results in a lightweight and computationally inexpensive network. Furthermore, depthwise dilated convolution in DDSC layer effectively expands the field of view of filters, allowing them to incorporate a larger context. The second subnetwork is a hierarchical multi-scale feature fusion (HMFF) module that uses parallel multi-resolution branches architecture to process the input feature map in order to extract the multi-scale feature information of the input image. Experimental results on the CIFAR-10, Malaria, and KvasirV1 datasets demonstrate that the proposed method is efficient, reducing the network parameters and computational cost by 65.02% and 39.78%, respectively, while maintaining the network performance compared to the MobileNetV1 baseline.展开更多
According to the basic theory on autofrettage and according to the 4th strength theory, several parameters and their relations are studied under ideal condition, including σej/σy, the equivalent stress of total stre...According to the basic theory on autofrettage and according to the 4th strength theory, several parameters and their relations are studied under ideal condition, including σej/σy, the equivalent stress of total stresses at elastoplastic juncture; σei/σy, the equivalent stress of total stresses at inside surface; σej'/σy, the equivalent stress of residual stresses at elastoplastic juncture; σei'/σy, the equivalent stress of residual stresses at inside surface; and p/σy, load-bearing capacity of an autofrettaged cylinder. By theoretical study on relations between the parameters, noticeable results and laws are achieved: to satisfy |σei'|=σy. the relation between kj and k is, k^2lnkj^2-k^2-kj^2+2=0, when k→∞, kj = √e = 1.648 72, as based on the 3rd strength theory, where k is the outside/inside radius ratio of a cylinder, kj is the ratio of elastoplastic juncture radius to inside radius of a cylinder; If the plastic region covers the whole wall of a cylinder, for compressive yield not to occur after removing autofrettage pressure, the ultimate k is k=-2.218 46 as based on the 3rd strength theory; With k=2.218 46, a cylinder's ultimate load-bearing capacity equals its entire yield pressure, or p/σy=21nk/√3; The maximum and optimum load-bearing capacity of an autofrettaged cylinder is just 2 times the loading which an unautofrettaged cylinder can bear elastically, or p/σy=2(k^2-1)/√3 k^2, and the limit of the load-bearing capacity of an autofrettaged cylinder is also just 2 times that of an unautofrettaged cylinder. The conclusions are the same as based on the 3rd strength theory, but some equations are different from each other.展开更多
Autofrettage is an effective measure to even distribution of stresses and raise load-bearing capacity for (ultra-)high pressure apparatus. Currently, the research on autofrettage has focused mostly on specific engin...Autofrettage is an effective measure to even distribution of stresses and raise load-bearing capacity for (ultra-)high pressure apparatus. Currently, the research on autofrettage has focused mostly on specific engineering problems, while general theoretical study is rarely done. To discover the general law contained in autofrettage theory, by the aid of the authors’ previous work and according to the third strength theory, theoretical problems about autofrettage are studied including residual stresses and their equivalent stress, total stresses and their equivalent stress, etc. Because of the equation of optimum depth of plastic zone which is presented in the authors’ previous work, the equations for the residual stresses and their equivalent stress as well as the total stress and their equivalent stress are simplified greatly. Thus the law of distribution of the residual stresses and their equivalent stress as well as the total stress and their equivalent stress and the varying tendency of these stresses are discovered. The relation among various parameters are revealed. The safe and optimum load-bearing conditions for cylinders are obtained. According to the results obtained by theoretical analysis, it is shown that if the two parameters, namely ratio of outside to inside radius, k, and depth of plastic zone, kj, meet the equation of optimum depth of plastic zone, when the pressure contained in an autofrettaged cylinder is lower than two times the initial yield pressure of the unautofrettaged cylinder, the equivalent residual stress and the equivalent total stress at the inside surface as well as the elastic-plastic juncture of a cylinder are lower than yield strength. When an autofrettaged cylinder is subjected to just two times the initial yield pressure of the unautofrettaged cylinder, the equivalent total stress within the whole plastic zone is just identically equal to the yield strength, or it is a constant. The proposed research theoretically depicts the stress state of ultra-)high pressure autofrettaged cylinder more accurately and more reasonably and provides the reference for design of (ultra-)high pressure apparatus.展开更多
The defence sector is now at an advanced level,catering to the global scenario,and countries also invest heavily in research and development.Countries around the world have spent a lot of money on research and develop...The defence sector is now at an advanced level,catering to the global scenario,and countries also invest heavily in research and development.Countries around the world have spent a lot of money on research and development over the years in order to stay ahead of their competitors.Lightweight materials are critical in defence applications because they allow components to be lighter without sacrificing strength.This review provides an overview of the research related to defence applications.The book provides comprehensive details on current trends in the application of lightweight materials in defence.This review also includes historical and current perspectives on defence technologies.It discusses uses of lightweight materials such as metal matrix composites,polymer composites,ceramic matrix composites,fiber composites in defence sectors Finally,the review paper also emphasizes future military applications of lightweight materials.展开更多
The lightweight shielding design of small reactors is a popular research topic.Based on a small helium-xenon-cooled solid reactor,the effects of neutron and photon shielding sequence and the number of shielding layers...The lightweight shielding design of small reactors is a popular research topic.Based on a small helium-xenon-cooled solid reactor,the effects of neutron and photon shielding sequence and the number of shielding layers on the radiation dose were first studied.It was found that when photons were shielded first and the number of shielding layers was odd,the radiation dose could be significantly reduced.To reduce the weight of the shielding body,the relative thickness of the shielding layers was optimized using the genetic algorithm.The optimized scheme can reduce the radiation dose by up to 57%and reduce the weight by 11.84%.To determine the total thickness of the shielding layers and avoid the local optimal solution of the genetic algorithm,a series of formulas that describes the relationship between the total thickness and the radiation dose was developed through large-scale calculations.A semi-empirical and semi-quantitative lightweight shielding design method is proposed to integrate the above shielding optimization method that verified by the Monte Carlo method.Finally,a code,SDIC1.0,was developed to achieve the optimized lightweight shielding design for small reactors.It was verified that the difference between the SDIC1.0 and the RMC code is approximately 10%and that the computation time is shortened by 6.3 times.展开更多
Significant progress has been made in computational imaging(CI),in which deep convolutional neural networks(CNNs)have demonstrated that sparse speckle patterns can be reconstructed.However,due to the limited“local”k...Significant progress has been made in computational imaging(CI),in which deep convolutional neural networks(CNNs)have demonstrated that sparse speckle patterns can be reconstructed.However,due to the limited“local”kernel size of the convolutional operator,for the spatially dense patterns,such as the generic face images,the performance of CNNs is limited.Here,we propose a“non-local”model,termed the Speckle-Transformer(SpT)UNet,for speckle feature extraction of generic face images.It is worth noting that the lightweight SpT UNet reveals a high efficiency and strong comparative performance with Pearson Correlation Coefficient(PCC),and structural similarity measure(SSIM)exceeding 0.989,and 0.950,respectively.展开更多
To better improve the lightweight and fatigue durability performance of the tractor cab,a multi-objective lightweight design of the cab was carried out in this study.First,the finite element model of the cab with coun...To better improve the lightweight and fatigue durability performance of the tractor cab,a multi-objective lightweight design of the cab was carried out in this study.First,the finite element model of the cab with counterweight loading was established and then confirmed by the physical testing,and use the inertial reliefmethod to obtain stress distribution under unit load.The cab-frame rigid-flexible couplingmulti-body dynamicsmodelwas built by Adams/car software.Taking the cab airbag mount displacement and acceleration signals acquired on the proving ground as the desired signals and obtaining the fatigue analysis load spectrum through Femfat-Lab virtual iteration.The fatigue simulation analysis is performed in nCode based on the Miner linear fatigue cumulative damage theory.Then,with themass and fatigue damage values as the optimization objectives,the bending-torsional stiffness and first-order bending-torsional mode as constraints,the thickness variables are screed based on the sensitivity analysis.The experimental design was carried out using the Optimal Latin hypercube method,and the multi-objective optimal design of the cab was carried out using theKriging approximationmodel fitting and particle swarmalgorithm.The weight of the optimized cab is reduced by 7.8%on the basis of meeting the fatigue durability performance.Finally,a seven-axis road simulation test rig was designed to verify its fatigue durability.The results show the optimized cab can consider both lightweight and durability.展开更多
The diagnosis of COVID-19 requires chest computed tomography(CT).High-resolution CT images can provide more diagnostic information to help doctors better diagnose the disease,so it is of clinical importance to study s...The diagnosis of COVID-19 requires chest computed tomography(CT).High-resolution CT images can provide more diagnostic information to help doctors better diagnose the disease,so it is of clinical importance to study super-resolution(SR)algorithms applied to CT images to improve the reso-lution of CT images.However,most of the existing SR algorithms are studied based on natural images,which are not suitable for medical images;and most of these algorithms improve the reconstruction quality by increasing the network depth,which is not suitable for machines with limited resources.To alleviate these issues,we propose a residual feature attentional fusion network for lightweight chest CT image super-resolution(RFAFN).Specifically,we design a contextual feature extraction block(CFEB)that can extract CT image features more efficiently and accurately than ordinary residual blocks.In addition,we propose a feature-weighted cascading strategy(FWCS)based on attentional feature fusion blocks(AFFB)to utilize the high-frequency detail information extracted by CFEB as much as possible via selectively fusing adjacent level feature information.Finally,we suggest a global hierarchical feature fusion strategy(GHFFS),which can utilize the hierarchical features more effectively than dense concatenation by progressively aggregating the feature information at various levels.Numerous experiments show that our method performs better than most of the state-of-the-art(SOTA)methods on the COVID-19 chest CT dataset.In detail,the peak signal-to-noise ratio(PSNR)is 0.11 dB and 0.47 dB higher on CTtest1 and CTtest2 at×3 SR compared to the suboptimal method,but the number of parameters and multi-adds are reduced by 22K and 0.43G,respectively.Our method can better recover chest CT image quality with fewer computational resources and effectively assist in COVID-19.展开更多
In response to the problem of the high cost and low efficiency of traditional water surface litter cleanup through manpower,a lightweight water surface litter detection algorithm based on improved YOLOv5s is proposed ...In response to the problem of the high cost and low efficiency of traditional water surface litter cleanup through manpower,a lightweight water surface litter detection algorithm based on improved YOLOv5s is proposed to provide core technical support for real-time water surface litter detection by water surface litter cleanup vessels.The method reduces network parameters by introducing the deep separable convolution GhostConv in the lightweight network GhostNet to substitute the ordinary convolution in the original YOLOv5s feature extraction and fusion network;introducing the C3Ghost module to substitute the C3 module in the original backbone and neck networks to further reduce computational effort.Using a Convolutional Block Attention Mechanism(CBAM)module in the backbone network to strengthen the network’s ability to extract significant target features from images.Finally,the loss function is optimized using the Focal-EIoU loss func-tion to improve the convergence speed and model accuracy.The experimental results illustrate that the improved algorithm outperforms the original Yolov5s in all aspects of the homemade water surface litter dataset and has certain advantages over some current mainstream algorithms in terms of model size,detection accuracy,and speed,which can deal with the problems of real-time detection of water surface litter in real life.展开更多
The importance of implantable biomaterials is growing up in recent days for modern medicine,especially fixation,replacement,and regeneration of load-bearing bones.Through the past several years,metals,ceramics,polymer...The importance of implantable biomaterials is growing up in recent days for modern medicine,especially fixation,replacement,and regeneration of load-bearing bones.Through the past several years,metals,ceramics,polymers,and their composites,have been used for the reconstruction of hard tissues.Special standards such as adequate mechanical and biocompatible properties are required to avoid rejection reactions of the tissues.Recently,a number of novel advanced biomaterials are developed as promising candidates.Amongst those,cerium-based biomaterials acquired attention as a substitution material for hard tissues reconstruction because of cerium antioxidative properties,which enabled it to be used to decrease mediators of inflammation.In addition,the eminent mechanical properties,as well as the perfect chemical and biological compatibilities,make cerium-based biomaterials attractive for biomedical application.展开更多
Encryption algorithms are one of the methods to protect dataduring its transmission through an unsafe transmission medium. But encryptionmethods need a lot of time during encryption and decryption, so itis necessary t...Encryption algorithms are one of the methods to protect dataduring its transmission through an unsafe transmission medium. But encryptionmethods need a lot of time during encryption and decryption, so itis necessary to find encryption algorithms that consume little time whilepreserving the security of the data. In this paper, more than one algorithmwas combined to obtain high security with a short implementation time. Achaotic system, DNA computing, and Salsa20 were combined. A proposed5D chaos system was used to generate more robust keys in a Salsa algorithmand DNA computing. Also, the confusion is performed using a new SBox.The proposed chaos system achieves three positive Lyapunov values.So results demonstrate of the proposed scheme has a sufficient peak signalto-noise ratio, a low correlation, and a large key space. These factors makeit more efficient than its classical counterpart and can resist statistical anddifferential attacks. The number of changing pixel rates (NPCR) and theunified averaged changed intensity (UACI) values were 0.99710 and UACI33.68. The entropy oscillates from 7.9965 to 7.9982 for the tested encryptedimages. The suggested approach is resistant to heavy attacks and takes lesstime to execute than previously discussed methods, making it an efficient,lightweight image encryption scheme. The method provides lower correlationcoefficients than other methods, another indicator of an efficient imageencryption system. Even though the proposed scheme has useful applicationsin image transmission, it still requires profound improvement in implementingthe high-intelligence scheme and verifying its feasibility on devices with theInternet of Things (IoT) enabled.展开更多
To solve the problems in online target detection on the embedded platform,such as high missed detection rate,low accuracy,and slow speed,a lightweight target recognition method of MobileNetV3-CenterNet is proposed.Thi...To solve the problems in online target detection on the embedded platform,such as high missed detection rate,low accuracy,and slow speed,a lightweight target recognition method of MobileNetV3-CenterNet is proposed.This method combines the anchor-free Centernet network with the MobileNetV3 small network and is trained on the University at Albany Detection and Tracking(UA-DETRAC)and the Pattern Analysis,Statical Modeling and Computational Learn-ing Visual Object Classes(PASCAL VOC)07+12 standard datasets.While reducing the scale of the network model,the MobileNetV3-CenterNet model shows a good balance in the accuracy and speed of target recognition and effectively solves the problems of missing detection of dense and small targets in online detection.To verify the recognition performance of the model,it is tested on 2683 images of the UA-DETRAC and PASCAL VOC 07+12 datasets,and compared with the analysis results of CenterNet-Deep Layer Aggregation(DLA)34,CenterNet-Residual Network(ResNet)18,CenterNet-MobileNetV3-large,You Only Look Once vision 3(YOLOv3),MobileNetV2-YOLOv3,Single Shot Multibox Detector(SSD),MobileNetV2-SSD and Faster region convolutional neural network(RCNN)models.The results show that the MobileNetV3-CenterNet model accurately rec-ognized the dense targets and small targets missed by other methods,and obtained a recognition accuracy of 99.4%with a running speed at 53(on a server)and 14(on an ipad)frame/s,respectively.The MobileNetV3-CenterNet lightweight target recognition method will provide effective technical support for the target recognition of deep learning networks in embedded platforms and online detection.展开更多
基金the National Natural Science Foundation of China(Grant Number 62076246).
文摘Human pose estimation aims to localize the body joints from image or video data.With the development of deeplearning,pose estimation has become a hot research topic in the field of computer vision.In recent years,humanpose estimation has achieved great success in multiple fields such as animation and sports.However,to obtainaccurate positioning results,existing methods may suffer from large model sizes,a high number of parameters,and increased complexity,leading to high computing costs.In this paper,we propose a new lightweight featureencoder to construct a high-resolution network that reduces the number of parameters and lowers the computingcost.We also introduced a semantic enhancement module that improves global feature extraction and networkperformance by combining channel and spatial dimensions.Furthermore,we propose a dense connected spatialpyramid pooling module to compensate for the decrease in image resolution and information loss in the network.Finally,ourmethod effectively reduces the number of parameters and complexitywhile ensuring high performance.Extensive experiments show that our method achieves a competitive performance while dramatically reducing thenumber of parameters,and operational complexity.Specifically,our method can obtain 89.9%AP score on MPIIVAL,while the number of parameters and the complexity of operations were reduced by 41%and 36%,respectively.
基金supported by the Korea Institute for Advancement of Technology(KIAT)Grant funded by theKorean Government(MOTIE)(P0008703,The Competency Development Program for Industry Specialists)MSIT under the ICAN(ICT Challenge and Advanced Network of HRD)Program(No.IITP-2022-RS-2022-00156310)supervised by the Institute of Information&Communication Technology Planning and Evaluation(IITP).
文摘With the advancement of wireless network technology,vast amounts of traffic have been generated,and malicious traffic attacks that threaten the network environment are becoming increasingly sophisticated.While signature-based detection methods,static analysis,and dynamic analysis techniques have been previously explored for malicious traffic detection,they have limitations in identifying diversified malware traffic patterns.Recent research has been focused on the application of machine learning to detect these patterns.However,applying machine learning to lightweight devices like IoT devices is challenging because of the high computational demands and complexity involved in the learning process.In this study,we examined methods for effectively utilizing machine learning-based malicious traffic detection approaches for lightweight devices.We introduced the suboptimal feature selection model(SFSM),a feature selection technique designed to reduce complexity while maintaining the effectiveness of malicious traffic detection.Detection performance was evaluated on various malicious traffic,benign,exploits,and generic,using the UNSW-NB15 dataset and SFSM sub-optimized hyperparameters for feature selection and narrowed the search scope to encompass all features.SFSM improved learning performance while minimizing complexity by considering feature selection and exhaustive search as two steps,a problem not considered in conventional models.Our experimental results showed that the detection accuracy was improved by approximately 20%compared to the random model,and the reduction in accuracy compared to the greedy model,which performs an exhaustive search on all features,was kept within 6%.Additionally,latency and complexity were reduced by approximately 96%and 99.78%,respectively,compared to the greedy model.This study demonstrates that malicious traffic can be effectively detected even in lightweight device environments.SFSM verified the possibility of detecting various attack traffic on lightweight devices.
基金The work is supported by the National Natural Science Foundation of China(Nos.U21A20124 and 52205059)the Key Research and Development Program of Zhejiang Province(No.2022C01039)。
文摘Galloping cheetahs,climbing mountain goats,and load hauling horses all show desirable locomotion capability,which motivates the development of quadruped robots.Among various quadruped robots,hydraulically driven quadruped robots show great potential in unstructured environments due to their discrete landing positions and large payloads.As the most critical movement unit of a quadruped robot,the limb leg unit(LLU)directly affects movement speed and reliability,and requires a compact and lightweight design.Inspired by the dexterous skeleton–muscle systems of cheetahs and humans,this paper proposes a highly integrated bionic actuator system for a better dynamic performance of an LLU.We propose that a cylinder barrel with multiple element interfaces and internal smooth channels is realized using metal additive manufacturing,and hybrid lattice structures are introduced into the lightweight design of the piston rod.In addition,additive manufacturing and topology optimization are incorporated to reduce the redundant material of the structural parts of the LLU.The mechanical properties of the actuator system are verified by numerical simulation and experiments,and the power density of the actuators is far greater than that of cheetah muscle.The mass of the optimized LLU is reduced by 24.5%,and the optimized LLU shows better response time performance when given a step signal,and presents a good trajectory tracking ability with the increase in motion frequency.
基金supported by project TRANSACT funded under H2020-EU.2.1.1.-INDUSTRIAL LEADERSHIP-Leadership in Enabling and Industrial Technologies-Information and Communication Technologies(Grant Agreement ID:101007260).
文摘The widespread and growing interest in the Internet of Things(IoT)may be attributed to its usefulness in many different fields.Physical settings are probed for data,which is then transferred via linked networks.There are several hurdles to overcome when putting IoT into practice,from managing server infrastructure to coordinating the use of tiny sensors.When it comes to deploying IoT,everyone agrees that security is the biggest issue.This is due to the fact that a large number of IoT devices exist in the physicalworld and thatmany of themhave constrained resources such as electricity,memory,processing power,and square footage.This research intends to analyse resource-constrained IoT devices,including RFID tags,sensors,and smart cards,and the issues involved with protecting them in such restricted circumstances.Using lightweight cryptography,the information sent between these gadgets may be secured.In order to provide a holistic picture,this research evaluates and contrasts well-known algorithms based on their implementation cost,hardware/software efficiency,and attack resistance features.We also emphasised how essential lightweight encryption is for striking a good cost-to-performance-to-security ratio.
基金supported by the National Natural Science Foundation of China(No.62176034)the Science and Technology Research Program of Chongqing Municipal Education Commission(No.KJZD-M202300604)the Natural Science Foundation of Chongqing(Nos.cstc2021jcyj-msxmX0518,2023NSCQ-MSX1781).
文摘Automatic crack detection of cement pavement chiefly benefits from the rapid development of deep learning,with convolutional neural networks(CNN)playing an important role in this field.However,as the performance of crack detection in cement pavement improves,the depth and width of the network structure are significantly increased,which necessitates more computing power and storage space.This limitation hampers the practical implementation of crack detection models on various platforms,particularly portable devices like small mobile devices.To solve these problems,we propose a dual-encoder-based network architecture that focuses on extracting more comprehensive fracture feature information and combines cross-fusion modules and coordinated attention mechanisms formore efficient feature fusion.Firstly,we use small channel convolution to construct shallow feature extractionmodule(SFEM)to extract low-level feature information of cracks in cement pavement images,in order to obtainmore information about cracks in the shallowfeatures of images.In addition,we construct large kernel atrous convolution(LKAC)to enhance crack information,which incorporates coordination attention mechanism for non-crack information filtering,and large kernel atrous convolution with different cores,using different receptive fields to extract more detailed edge and context information.Finally,the three-stage feature map outputs from the shallow feature extraction module is cross-fused with the two-stage feature map outputs from the large kernel atrous convolution module,and the shallow feature and detailed edge feature are fully fused to obtain the final crack prediction map.We evaluate our method on three public crack datasets:DeepCrack,CFD,and Crack500.Experimental results on theDeepCrack dataset demonstrate the effectiveness of our proposed method compared to state-of-the-art crack detection methods,which achieves Precision(P)87.2%,Recall(R)87.7%,and F-score(F1)87.4%.Thanks to our lightweight crack detectionmodel,the parameter count of the model in real-world detection scenarios has been significantly reduced to less than 2M.This advancement also facilitates technical support for portable scene detection.
文摘Blockchain technology has garnered significant attention from global organizations and researchers due to its potential as a solution for centralized system challenges.Concurrently,the Internet of Things(IoT)has revolutionized the Fourth Industrial Revolution by enabling interconnected devices to offer innovative services,ultimately enhancing human lives.This paper presents a new approach utilizing lightweight blockchain technology,effectively reducing the computational burden typically associated with conventional blockchain systems.By integrating this lightweight blockchain with IoT systems,substantial reductions in implementation time and computational complexity can be achieved.Moreover,the paper proposes the utilization of the Okamoto Uchiyama encryption algorithm,renowned for its homomorphic characteristics,to reinforce the privacy and security of IoT-generated data.The integration of homomorphic encryption and blockchain technology establishes a secure and decentralized platformfor storing and analyzing sensitive data of the supply chain data.This platformfacilitates the development of some business models and empowers decentralized applications to perform computations on encrypted data while maintaining data privacy.The results validate the robust security of the proposed system,comparable to standard blockchain implementations,leveraging the distinctive homomorphic attributes of the Okamoto Uchiyama algorithm and the lightweight blockchain paradigm.
文摘With the growth of the Internet,more and more business is being done online,for example,online offices,online education and so on.While this makes people’s lives more convenient,it also increases the risk of the network being attacked by malicious code.Therefore,it is important to identify malicious codes on computer systems efficiently.However,most of the existing malicious code detection methods have two problems:(1)The ability of the model to extract features is weak,resulting in poor model performance.(2)The large scale of model data leads to difficulties deploying on devices with limited resources.Therefore,this paper proposes a lightweight malicious code identification model Lightweight Malicious Code Classification Method Based on Improved SqueezeNet(LCMISNet).In this paper,the MFire lightweight feature extraction module is constructed by proposing a feature slicing module and a multi-size depthwise separable convolution module.The feature slicing module reduces the number of parameters by grouping features.The multi-size depthwise separable convolution module reduces the number of parameters and enhances the feature extraction capability by replacing the standard convolution with depthwise separable convolution with different convolution kernel sizes.In addition,this paper also proposes a feature splicing module to connect the MFire lightweight feature extraction module based on the feature reuse and constructs the lightweight model LCMISNet.The malicious code recognition accuracy of LCMISNet on the BIG 2015 dataset and the Malimg dataset reaches 98.90% and 99.58%,respectively.It proves that LCMISNet has a powerful malicious code recognition performance.In addition,compared with other network models,LCMISNet has better performance,and a lower number of parameters and computations.
文摘A lightweight malware detection and family classification system for the Internet of Things (IoT) was designed to solve the difficulty of deploying defense models caused by the limited computing and storage resources of IoT devices. By training complex models with IoT software gray-scale images and utilizing the gradient-weighted class-activated mapping technique, the system can identify key codes that influence model decisions. This allows for the reconstruction of gray-scale images to train a lightweight model called LMDNet for malware detection. Additionally, the multi-teacher knowledge distillation method is employed to train KD-LMDNet, which focuses on classifying malware families. The results indicate that the model’s identification speed surpasses that of traditional methods by 23.68%. Moreover, the accuracy achieved on the Malimg dataset for family classification is an impressive 99.07%. Furthermore, with a model size of only 0.45M, it appears to be well-suited for the IoT environment. By training complex models using IoT software gray-scale images and utilizing the gradient-weighted class-activated mapping technique, the system can identify key codes that influence model decisions. This allows for the reconstruction of gray-scale images to train a lightweight model called LMDNet for malware detection. Thus, the presented approach can address the challenges associated with malware detection and family classification in IoT devices.
文摘Convolutional neural networks (CNNs) are widely used in image classification tasks, but their increasing model size and computation make them challenging to implement on embedded systems with constrained hardware resources. To address this issue, the MobileNetV1 network was developed, which employs depthwise convolution to reduce network complexity. MobileNetV1 employs a stride of 2 in several convolutional layers to decrease the spatial resolution of feature maps, thereby lowering computational costs. However, this stride setting can lead to a loss of spatial information, particularly affecting the detection and representation of smaller objects or finer details in images. To maintain the trade-off between complexity and model performance, a lightweight convolutional neural network with hierarchical multi-scale feature fusion based on the MobileNetV1 network is proposed. The network consists of two main subnetworks. The first subnetwork uses a depthwise dilated separable convolution (DDSC) layer to learn imaging features with fewer parameters, which results in a lightweight and computationally inexpensive network. Furthermore, depthwise dilated convolution in DDSC layer effectively expands the field of view of filters, allowing them to incorporate a larger context. The second subnetwork is a hierarchical multi-scale feature fusion (HMFF) module that uses parallel multi-resolution branches architecture to process the input feature map in order to extract the multi-scale feature information of the input image. Experimental results on the CIFAR-10, Malaria, and KvasirV1 datasets demonstrate that the proposed method is efficient, reducing the network parameters and computational cost by 65.02% and 39.78%, respectively, while maintaining the network performance compared to the MobileNetV1 baseline.
文摘According to the basic theory on autofrettage and according to the 4th strength theory, several parameters and their relations are studied under ideal condition, including σej/σy, the equivalent stress of total stresses at elastoplastic juncture; σei/σy, the equivalent stress of total stresses at inside surface; σej'/σy, the equivalent stress of residual stresses at elastoplastic juncture; σei'/σy, the equivalent stress of residual stresses at inside surface; and p/σy, load-bearing capacity of an autofrettaged cylinder. By theoretical study on relations between the parameters, noticeable results and laws are achieved: to satisfy |σei'|=σy. the relation between kj and k is, k^2lnkj^2-k^2-kj^2+2=0, when k→∞, kj = √e = 1.648 72, as based on the 3rd strength theory, where k is the outside/inside radius ratio of a cylinder, kj is the ratio of elastoplastic juncture radius to inside radius of a cylinder; If the plastic region covers the whole wall of a cylinder, for compressive yield not to occur after removing autofrettage pressure, the ultimate k is k=-2.218 46 as based on the 3rd strength theory; With k=2.218 46, a cylinder's ultimate load-bearing capacity equals its entire yield pressure, or p/σy=21nk/√3; The maximum and optimum load-bearing capacity of an autofrettaged cylinder is just 2 times the loading which an unautofrettaged cylinder can bear elastically, or p/σy=2(k^2-1)/√3 k^2, and the limit of the load-bearing capacity of an autofrettaged cylinder is also just 2 times that of an unautofrettaged cylinder. The conclusions are the same as based on the 3rd strength theory, but some equations are different from each other.
基金supported by Scientific Research Fund of Hunan Provincial Education Department(Grant No. 12A087)Innovation Fund for Technology Based Firms(Grant No. 09C26214305047)
文摘Autofrettage is an effective measure to even distribution of stresses and raise load-bearing capacity for (ultra-)high pressure apparatus. Currently, the research on autofrettage has focused mostly on specific engineering problems, while general theoretical study is rarely done. To discover the general law contained in autofrettage theory, by the aid of the authors’ previous work and according to the third strength theory, theoretical problems about autofrettage are studied including residual stresses and their equivalent stress, total stresses and their equivalent stress, etc. Because of the equation of optimum depth of plastic zone which is presented in the authors’ previous work, the equations for the residual stresses and their equivalent stress as well as the total stress and their equivalent stress are simplified greatly. Thus the law of distribution of the residual stresses and their equivalent stress as well as the total stress and their equivalent stress and the varying tendency of these stresses are discovered. The relation among various parameters are revealed. The safe and optimum load-bearing conditions for cylinders are obtained. According to the results obtained by theoretical analysis, it is shown that if the two parameters, namely ratio of outside to inside radius, k, and depth of plastic zone, kj, meet the equation of optimum depth of plastic zone, when the pressure contained in an autofrettaged cylinder is lower than two times the initial yield pressure of the unautofrettaged cylinder, the equivalent residual stress and the equivalent total stress at the inside surface as well as the elastic-plastic juncture of a cylinder are lower than yield strength. When an autofrettaged cylinder is subjected to just two times the initial yield pressure of the unautofrettaged cylinder, the equivalent total stress within the whole plastic zone is just identically equal to the yield strength, or it is a constant. The proposed research theoretically depicts the stress state of ultra-)high pressure autofrettaged cylinder more accurately and more reasonably and provides the reference for design of (ultra-)high pressure apparatus.
文摘The defence sector is now at an advanced level,catering to the global scenario,and countries also invest heavily in research and development.Countries around the world have spent a lot of money on research and development over the years in order to stay ahead of their competitors.Lightweight materials are critical in defence applications because they allow components to be lighter without sacrificing strength.This review provides an overview of the research related to defence applications.The book provides comprehensive details on current trends in the application of lightweight materials in defence.This review also includes historical and current perspectives on defence technologies.It discusses uses of lightweight materials such as metal matrix composites,polymer composites,ceramic matrix composites,fiber composites in defence sectors Finally,the review paper also emphasizes future military applications of lightweight materials.
基金sponsored by Natural Science Foundation of Shanghai (No.22ZR1431900)the Young Elite Scientist Sponsorship Program of the China National Nuclear Corporation (CNNC).
文摘The lightweight shielding design of small reactors is a popular research topic.Based on a small helium-xenon-cooled solid reactor,the effects of neutron and photon shielding sequence and the number of shielding layers on the radiation dose were first studied.It was found that when photons were shielded first and the number of shielding layers was odd,the radiation dose could be significantly reduced.To reduce the weight of the shielding body,the relative thickness of the shielding layers was optimized using the genetic algorithm.The optimized scheme can reduce the radiation dose by up to 57%and reduce the weight by 11.84%.To determine the total thickness of the shielding layers and avoid the local optimal solution of the genetic algorithm,a series of formulas that describes the relationship between the total thickness and the radiation dose was developed through large-scale calculations.A semi-empirical and semi-quantitative lightweight shielding design method is proposed to integrate the above shielding optimization method that verified by the Monte Carlo method.Finally,a code,SDIC1.0,was developed to achieve the optimized lightweight shielding design for small reactors.It was verified that the difference between the SDIC1.0 and the RMC code is approximately 10%and that the computation time is shortened by 6.3 times.
基金funding support from the Science and Technology Commission of Shanghai Municipality(Grant No.21DZ1100500)the Shanghai Frontiers Science Center Program(2021-2025 No.20)+2 种基金the Zhangjiang National Innovation Demonstration Zone(Grant No.ZJ2019ZD-005)supported by a fellowship from the China Postdoctoral Science Foundation(2020M671169)the International Postdoctoral Exchange Program from the Administrative Committee of Post-Doctoral Researchers of China([2020]33)。
文摘Significant progress has been made in computational imaging(CI),in which deep convolutional neural networks(CNNs)have demonstrated that sparse speckle patterns can be reconstructed.However,due to the limited“local”kernel size of the convolutional operator,for the spatially dense patterns,such as the generic face images,the performance of CNNs is limited.Here,we propose a“non-local”model,termed the Speckle-Transformer(SpT)UNet,for speckle feature extraction of generic face images.It is worth noting that the lightweight SpT UNet reveals a high efficiency and strong comparative performance with Pearson Correlation Coefficient(PCC),and structural similarity measure(SSIM)exceeding 0.989,and 0.950,respectively.
基金supported in part by the Science and Technology Major Project of Guangxi under Grants AA18242033 and AA19182004in part by the Key Research andDevelopment Program of Guangxi AB21196029+3 种基金in part by the Scientific Research and Technology Development in Liuzhou 2020GAAA0404,2021AAA0104 and 2021AAA0112in part by the Guangxi Higher Education Undergraduate Teaching Reform Project Grant 2021JGA180in part by the GUET Education Undergraduate Teaching Reform Project Grant JGB202002in part by the Innovation Project of GUET Graduate Education (2022YCXS017).
文摘To better improve the lightweight and fatigue durability performance of the tractor cab,a multi-objective lightweight design of the cab was carried out in this study.First,the finite element model of the cab with counterweight loading was established and then confirmed by the physical testing,and use the inertial reliefmethod to obtain stress distribution under unit load.The cab-frame rigid-flexible couplingmulti-body dynamicsmodelwas built by Adams/car software.Taking the cab airbag mount displacement and acceleration signals acquired on the proving ground as the desired signals and obtaining the fatigue analysis load spectrum through Femfat-Lab virtual iteration.The fatigue simulation analysis is performed in nCode based on the Miner linear fatigue cumulative damage theory.Then,with themass and fatigue damage values as the optimization objectives,the bending-torsional stiffness and first-order bending-torsional mode as constraints,the thickness variables are screed based on the sensitivity analysis.The experimental design was carried out using the Optimal Latin hypercube method,and the multi-objective optimal design of the cab was carried out using theKriging approximationmodel fitting and particle swarmalgorithm.The weight of the optimized cab is reduced by 7.8%on the basis of meeting the fatigue durability performance.Finally,a seven-axis road simulation test rig was designed to verify its fatigue durability.The results show the optimized cab can consider both lightweight and durability.
基金supported by the General Project of Natural Science Foundation of Hebei Province of China(H2019201378)the Foundation of the President of Hebei University(XZJJ201917)the Special Project for Cultivating Scientific and Technological Innovation Ability of University and Middle School Students of Hebei Province(2021H060306).
文摘The diagnosis of COVID-19 requires chest computed tomography(CT).High-resolution CT images can provide more diagnostic information to help doctors better diagnose the disease,so it is of clinical importance to study super-resolution(SR)algorithms applied to CT images to improve the reso-lution of CT images.However,most of the existing SR algorithms are studied based on natural images,which are not suitable for medical images;and most of these algorithms improve the reconstruction quality by increasing the network depth,which is not suitable for machines with limited resources.To alleviate these issues,we propose a residual feature attentional fusion network for lightweight chest CT image super-resolution(RFAFN).Specifically,we design a contextual feature extraction block(CFEB)that can extract CT image features more efficiently and accurately than ordinary residual blocks.In addition,we propose a feature-weighted cascading strategy(FWCS)based on attentional feature fusion blocks(AFFB)to utilize the high-frequency detail information extracted by CFEB as much as possible via selectively fusing adjacent level feature information.Finally,we suggest a global hierarchical feature fusion strategy(GHFFS),which can utilize the hierarchical features more effectively than dense concatenation by progressively aggregating the feature information at various levels.Numerous experiments show that our method performs better than most of the state-of-the-art(SOTA)methods on the COVID-19 chest CT dataset.In detail,the peak signal-to-noise ratio(PSNR)is 0.11 dB and 0.47 dB higher on CTtest1 and CTtest2 at×3 SR compared to the suboptimal method,but the number of parameters and multi-adds are reduced by 22K and 0.43G,respectively.Our method can better recover chest CT image quality with fewer computational resources and effectively assist in COVID-19.
基金Support for this work was in part from the China University Industry-University Research Innovation Fund Project(No.2022BL052),author B.T,https://www.cutech.edu.cnin part by the Science and Technology InnovationR&DProject of the State GeneralAdministration of Sports of China(No.22KJCX024),author B.T,https://www.sport.gov.cn+1 种基金in part by the Major Project of Philosophy and Social Science Research in Higher Education Institutions in Hubei Province(No.21ZD054),author B.T,https://jyt.hubei.gov.cnKey Project of Hubei Provincial Key Laboratory of Intelligent Transportation Technology and Equipment Open Fund(No.2022XZ106),author B.T,https://hbpu.edu.cn.
文摘In response to the problem of the high cost and low efficiency of traditional water surface litter cleanup through manpower,a lightweight water surface litter detection algorithm based on improved YOLOv5s is proposed to provide core technical support for real-time water surface litter detection by water surface litter cleanup vessels.The method reduces network parameters by introducing the deep separable convolution GhostConv in the lightweight network GhostNet to substitute the ordinary convolution in the original YOLOv5s feature extraction and fusion network;introducing the C3Ghost module to substitute the C3 module in the original backbone and neck networks to further reduce computational effort.Using a Convolutional Block Attention Mechanism(CBAM)module in the backbone network to strengthen the network’s ability to extract significant target features from images.Finally,the loss function is optimized using the Focal-EIoU loss func-tion to improve the convergence speed and model accuracy.The experimental results illustrate that the improved algorithm outperforms the original Yolov5s in all aspects of the homemade water surface litter dataset and has certain advantages over some current mainstream algorithms in terms of model size,detection accuracy,and speed,which can deal with the problems of real-time detection of water surface litter in real life.
文摘The importance of implantable biomaterials is growing up in recent days for modern medicine,especially fixation,replacement,and regeneration of load-bearing bones.Through the past several years,metals,ceramics,polymers,and their composites,have been used for the reconstruction of hard tissues.Special standards such as adequate mechanical and biocompatible properties are required to avoid rejection reactions of the tissues.Recently,a number of novel advanced biomaterials are developed as promising candidates.Amongst those,cerium-based biomaterials acquired attention as a substitution material for hard tissues reconstruction because of cerium antioxidative properties,which enabled it to be used to decrease mediators of inflammation.In addition,the eminent mechanical properties,as well as the perfect chemical and biological compatibilities,make cerium-based biomaterials attractive for biomedical application.
文摘Encryption algorithms are one of the methods to protect dataduring its transmission through an unsafe transmission medium. But encryptionmethods need a lot of time during encryption and decryption, so itis necessary to find encryption algorithms that consume little time whilepreserving the security of the data. In this paper, more than one algorithmwas combined to obtain high security with a short implementation time. Achaotic system, DNA computing, and Salsa20 were combined. A proposed5D chaos system was used to generate more robust keys in a Salsa algorithmand DNA computing. Also, the confusion is performed using a new SBox.The proposed chaos system achieves three positive Lyapunov values.So results demonstrate of the proposed scheme has a sufficient peak signalto-noise ratio, a low correlation, and a large key space. These factors makeit more efficient than its classical counterpart and can resist statistical anddifferential attacks. The number of changing pixel rates (NPCR) and theunified averaged changed intensity (UACI) values were 0.99710 and UACI33.68. The entropy oscillates from 7.9965 to 7.9982 for the tested encryptedimages. The suggested approach is resistant to heavy attacks and takes lesstime to execute than previously discussed methods, making it an efficient,lightweight image encryption scheme. The method provides lower correlationcoefficients than other methods, another indicator of an efficient imageencryption system. Even though the proposed scheme has useful applicationsin image transmission, it still requires profound improvement in implementingthe high-intelligence scheme and verifying its feasibility on devices with theInternet of Things (IoT) enabled.
基金supported by Research and Development Project of Key Core Technology and Common Technology in Shanxi Province(No.2020XXX009).
文摘To solve the problems in online target detection on the embedded platform,such as high missed detection rate,low accuracy,and slow speed,a lightweight target recognition method of MobileNetV3-CenterNet is proposed.This method combines the anchor-free Centernet network with the MobileNetV3 small network and is trained on the University at Albany Detection and Tracking(UA-DETRAC)and the Pattern Analysis,Statical Modeling and Computational Learn-ing Visual Object Classes(PASCAL VOC)07+12 standard datasets.While reducing the scale of the network model,the MobileNetV3-CenterNet model shows a good balance in the accuracy and speed of target recognition and effectively solves the problems of missing detection of dense and small targets in online detection.To verify the recognition performance of the model,it is tested on 2683 images of the UA-DETRAC and PASCAL VOC 07+12 datasets,and compared with the analysis results of CenterNet-Deep Layer Aggregation(DLA)34,CenterNet-Residual Network(ResNet)18,CenterNet-MobileNetV3-large,You Only Look Once vision 3(YOLOv3),MobileNetV2-YOLOv3,Single Shot Multibox Detector(SSD),MobileNetV2-SSD and Faster region convolutional neural network(RCNN)models.The results show that the MobileNetV3-CenterNet model accurately rec-ognized the dense targets and small targets missed by other methods,and obtained a recognition accuracy of 99.4%with a running speed at 53(on a server)and 14(on an ipad)frame/s,respectively.The MobileNetV3-CenterNet lightweight target recognition method will provide effective technical support for the target recognition of deep learning networks in embedded platforms and online detection.