Lightweight high/medium-entropy alloys(H/MEAs)possess attractive properties such as high strength-to-weight ratios,however,their limited room-temperature tensile ductility hinders their widespread engi-neering impleme...Lightweight high/medium-entropy alloys(H/MEAs)possess attractive properties such as high strength-to-weight ratios,however,their limited room-temperature tensile ductility hinders their widespread engi-neering implementation,for instance in aerospace structural components.This work achieved a transfor-mative improvement of room-temperature tensile ductility in Ti-V-Zr-Nb MEAs with densities of 5.4-6.5 g/cm3,via ingenious composition modulation.Through the systematic co-adjustment of Ti and V contents,an intrinsic ductility mechanism was unveiled,manifested by a transition from predominant intergranular brittle fracture to pervasive ductile dimpled rupture.Notably,the modulated deformation mechanisms evolved from solitary slip toward collaborative multiple slip modes,without significantly compromising strength.Compared to equimolar Ti-V-Zr-Nb,a(Ti1.5V)3ZrNb composition demonstrated an impressive 360%improvement in elongation while sustaining a high yield strength of around 800 MPa.Increasing Ti and V not only purified the grain boundaries by reducing detrimental phases,but also tai-lored the deformation dislocation configurations.These insights expanded the applicability of lightweight HEAs to areas demanding combined high strength and ductility.展开更多
The microstructural evolution,phase stability,and mechanical properties of Al-Li-Mg-Ti-M(M=Zn,Zr,V)lightweight high-entropy alloys(LW-HEAs)were investigated.The LW-HEAs with three components,Al_(20)Li_(20)Mg_(10)-Ti_(...The microstructural evolution,phase stability,and mechanical properties of Al-Li-Mg-Ti-M(M=Zn,Zr,V)lightweight high-entropy alloys(LW-HEAs)were investigated.The LW-HEAs with three components,Al_(20)Li_(20)Mg_(10)-Ti_(40)Zn_(10)(#Zn),Al_(20)Li_(20)Mg_(10)Ti_(30)Zr_(20)(#Zr),and Al_(20)Li_(20)Mg_(10)Ti_(30)V_(20)(#V),were designed according to the thermo-dynamic design criteria of HEA,and prepared via a combination process of mechanical alloying and cold-press sintering.The effects of alloy composition and sintering temperature on the microstructure and mechanical properties of the LW-HEAs were studied.The results show that the as-milled Al-Li-Mg-Ti-M(M=Zn,Zr,V)LW-HEAs form a simple structure with HCP-type solid solution as the primary phase,a dual-HCP type solid solution phase,and a BCC phase,respectively.After cold-press sintering,the#Zn and#V alloys undergo obvious phase transformation;while the#Zr alloy with dual-HCP phases exhibits the best phase stability during heat treatment.The#V-750°C alloy demonstrates the maximum hardness and specific strength of HV 595.2 and 625 MPa∙cm3/g,respectively,under the combined effect of solid solution strengthening of BCC phase and precipitation strengthening ofβ-AlTi_(3).Moreover,the#Zr-650°C,#Zr-750°C,and#Zn-650°C alloys are expected to have excellent plasticity.展开更多
Tomato plant diseases often first manifest on the leaves,making the detection of tomato leaf diseases particularly crucial for the tomato cultivation industry.However,conventional deep learning models face challenges ...Tomato plant diseases often first manifest on the leaves,making the detection of tomato leaf diseases particularly crucial for the tomato cultivation industry.However,conventional deep learning models face challenges such as large model sizes and slow detection speeds when deployed on resource-constrained platforms and agricultural machinery.This paper proposes a lightweight model for detecting tomato leaf diseases,named LT-YOLO,based on the YOLOv8n architecture.First,we enhance the C2f module into a RepViT Block(RVB)with decoupled token and channel mixers to reduce the cost of feature extraction.Next,we incorporate a novel Efficient Multi-Scale Attention(EMA)mechanism in the deeper layers of the backbone to improve detection of critical disease features.Additionally,we design a lightweight detection head,LT-Detect,using Partial Convolution(PConv)to significantly reduce the classification and localization costs during detection.Finally,we introduce a Receptive Field Block(RFB)in the shallow layers of the backbone to expand the model’s receptive field,enabling effective detection of diseases at various scales.The improved model reduces the number of parameters by 43%and the computational load by 50%.Additionally,it achieves a mean Average Precision(mAP)of 90.9%on a publicly available dataset containing 3641 images of tomato leaf diseases,with only a 0.7%decrease compared to the baseline model.This demonstrates that the model maintains excellent accuracy while being lightweight,making it suitable for rapid detection of tomato leaf diseases.展开更多
As batteries become increasingly essential for energy storage technologies,battery prognosis,and diagnosis remain central to ensure reliable operation and effective management,as well as to aid the in-depth investigat...As batteries become increasingly essential for energy storage technologies,battery prognosis,and diagnosis remain central to ensure reliable operation and effective management,as well as to aid the in-depth investigation of degradation mechanisms.However,dynamic operating conditions,cell-to-cell inconsistencies,and limited availability of labeled data have posed significant challenges to accurate and robust prognosis and diagnosis.Herein,we introduce a time-series-decomposition-based ensembled lightweight learning model(TELL-Me),which employs a synergistic dual-module framework to facilitate accurate and reliable forecasting.The feature module formulates features with physical implications and sheds light on battery aging mechanisms,while the gradient module monitors capacity degradation rates and captures aging trend.TELL-Me achieves high accuracy in end-of-life prediction using minimal historical data from a single battery without requiring offline training dataset,and demonstrates impressive generality and robustness across various operating conditions and battery types.Additionally,by correlating feature contributions with degradation mechanisms across different datasets,TELL-Me is endowed with the diagnostic ability that not only enhances prediction reliability but also provides critical insights into the design and optimization of next-generation batteries.展开更多
The Internet of Things(IoT)has gained substantial attention in both academic research and real-world applications.The proliferation of interconnected devices across various domains promises to deliver intelligent and ...The Internet of Things(IoT)has gained substantial attention in both academic research and real-world applications.The proliferation of interconnected devices across various domains promises to deliver intelligent and advanced services.However,this rapid expansion also heightens the vulnerability of the IoT ecosystem to security threats.Consequently,innovative solutions capable of effectively mitigating risks while accommodating the unique constraints of IoT environments are urgently needed.Recently,the convergence of Blockchain technology and IoT has introduced a decentralized and robust framework for securing data and interactions,commonly referred to as the Internet of Blockchained Things(IoBT).Extensive research efforts have been devoted to adapting Blockchain technology to meet the specific requirements of IoT deployments.Within this context,consensus algorithms play a critical role in assessing the feasibility of integrating Blockchain into IoT ecosystems.The adoption of efficient and lightweight consensus mechanisms for block validation has become increasingly essential.This paper presents a comprehensive examination of lightweight,constraint-aware consensus algorithms tailored for IoBT.The study categorizes these consensus mechanisms based on their core operations,the security of the block validation process,the incorporation of AI techniques,and the specific applications they are designed to support.展开更多
Introducing B2 ordering can effectively improve the mechanical properties of lightweight refractory high-entropy alloys(LRHEAs).However,(Zr,Al)-enriched B2 precipitates generally reduce the ductility because their ord...Introducing B2 ordering can effectively improve the mechanical properties of lightweight refractory high-entropy alloys(LRHEAs).However,(Zr,Al)-enriched B2 precipitates generally reduce the ductility because their ordering characteristic is destroyed after dislocation shearing.Meanwhile,the local chemical order(LCO)cannot provide an adequate strengthening effect due to its small size.展开更多
In recent years,with the development of synthetic aperture radar(SAR)technology and the widespread application of deep learning,lightweight detection of SAR images has emerged as a research direction.The ultimate goal...In recent years,with the development of synthetic aperture radar(SAR)technology and the widespread application of deep learning,lightweight detection of SAR images has emerged as a research direction.The ultimate goal is to reduce computational and storage requirements while ensuring detection accuracy and reliability,making it an ideal choice for achieving rapid response and efficient processing.In this regard,a lightweight SAR ship target detection algorithm based on YOLOv8 was proposed in this study.Firstly,the C2f-Sc module was designed by fusing the C2f in the backbone network with the ScConv to reduce spatial redundancy and channel redundancy between features in convolutional neural networks.At the same time,the Ghost module was introduced into the neck network to effectively reduce model parameters and computational complexity.A relatively lightweight EMA attention mechanism was added to the neck network to promote the effective fusion of features at different levels.Experimental results showed that the Parameters and GFLOPs of the improved model are reduced by 8.5%and 7.0%when mAP@0.5 and mAP@0.5:0.95 are increased by 0.7%and 1.8%,respectively.It makes the model lightweight and improves the detection accuracy,which has certain application value.展开更多
Semantic segmentation of eye images is a complex task with important applications in human–computer interaction,cognitive science,and neuroscience.Achieving real-time,accurate,and robust segmentation algorithms is cr...Semantic segmentation of eye images is a complex task with important applications in human–computer interaction,cognitive science,and neuroscience.Achieving real-time,accurate,and robust segmentation algorithms is crucial for computationally limited portable devices such as augmented reality and virtual reality.With the rapid advancements in deep learning,many network models have been developed specifically for eye image segmentation.Some methods divide the segmentation process into multiple stages to achieve model parameter miniaturization while enhancing output through post processing techniques to improve segmentation accuracy.These approaches significantly increase the inference time.Other networks adopt more complex encoding and decoding modules to achieve end-to-end output,which requires substantial computation.Therefore,balancing the model’s size,accuracy,and computational complexity is essential.To address these challenges,we propose a lightweight asymmetric UNet architecture and a projection loss function.We utilize ResNet-3 layer blocks to enhance feature extraction efficiency in the encoding stage.In the decoding stage,we employ regular convolutions and skip connections to upscale the feature maps from the latent space to the original image size,balancing the model size and segmentation accuracy.In addition,we leverage the geometric features of the eye region and design a projection loss function to further improve the segmentation accuracy without adding any additional inference computational cost.We validate our approach on the OpenEDS2019 dataset for virtual reality and achieve state-of-the-art performance with 95.33%mean intersection over union(mIoU).Our model has only 0.63M parameters and 350 FPS,which are 68%and 200%of the state-of-the-art model RITNet,respectively.展开更多
There is a problem of real-time detection difficulty in road surface damage detection. This paper proposes an improved lightweight model based on you only look once version 5(YOLOv5). Firstly, this paper fully utilize...There is a problem of real-time detection difficulty in road surface damage detection. This paper proposes an improved lightweight model based on you only look once version 5(YOLOv5). Firstly, this paper fully utilized the convolutional neural network(CNN) + ghosting bottleneck(G_bneck) architecture to reduce redundant feature maps. Afterwards, we upgraded the original upsampling algorithm to content-aware reassembly of features(CARAFE) and increased the receptive field. Finally, we replaced the spatial pyramid pooling fast(SPPF) module with the basic receptive field block(Basic RFB) pooling module and added dilated convolution. After comparative experiments, we can see that the number of parameters and model size of the improved algorithm in this paper have been reduced by nearly half compared to the YOLOv5s. The frame rate per second(FPS) has been increased by 3.25 times. The mean average precision(m AP@0.5: 0.95) has increased by 8%—17% compared to other lightweight algorithms.展开更多
Aiming at the problems of low detection efficiency and difficult positioning of traditional steel surface defect detection methods,a lightweight steel surface defect detection model based on you only look once version...Aiming at the problems of low detection efficiency and difficult positioning of traditional steel surface defect detection methods,a lightweight steel surface defect detection model based on you only look once version 7(YOLOv7)is proposed.First,a cascading style sheets(CSS)block module is proposed,which uses more lightweight operations to obtain redundant information in the feature map,reduces the amount of computation,and effectively improves the detection speed.Secondly,the improved spatial pyramid pooling with cross stage partial convolutions(SPPCSPC)structure is adopted to ensure that the model can also pay attention to the defect location information while predicting the defect category information,obtain richer defect features.In addition,the convolution operation in the original model is simplified,which significantly reduces the size of the model and helps to improve the detection speed.Finally,using efficient intersection over union(EIOU)loss to focus on high-quality anchors,speed up convergence and improve positioning accuracy.Experiments were carried out on the Northeastern University-defect(NEU-DET)steel surface defect dataset.Compared with the original YOLOv7 model,the number of parameters of this model was reduced by 40%,the frames per second(FPS)reached 112,and the average accuracy reached 79.1%,the detection accuracy and speed have been improved,which can meet the needs of steel surface defect detection.展开更多
Current you only look once(YOLO)-based algorithm model is facing the challenge of overwhelming parameters and calculation complexity under the printed circuit board(PCB)defect detection application scenario.In order t...Current you only look once(YOLO)-based algorithm model is facing the challenge of overwhelming parameters and calculation complexity under the printed circuit board(PCB)defect detection application scenario.In order to solve this problem,we propose a new method,which combined the lightweight network mobile vision transformer(Mobile Vi T)with the convolutional block attention module(CBAM)mechanism and the new regression loss function.This method needed less computation resources,making it more suitable for embedded edge detection devices.Meanwhile,the new loss function improved the positioning accuracy of the bounding box and enhanced the robustness of the model.In addition,experiments on public datasets demonstrate that the improved model achieves an average accuracy of 87.9%across six typical defect detection tasks,while reducing computational costs by nearly 90%.It significantly reduces the model's computational requirements while maintaining accuracy,ensuring reliable performance for edge deployment.展开更多
As the Internet of Medical Things (IoMT) continues to expand, smart health-monitoring devices generate vast amounts of valuable data while simultaneously raising critical security and privacy challenges. Blockchain te...As the Internet of Medical Things (IoMT) continues to expand, smart health-monitoring devices generate vast amounts of valuable data while simultaneously raising critical security and privacy challenges. Blockchain technology presents a promising avenue to address these concerns due to its inherent decentralization and security features. However, scalability remains a persistent hurdle, particularly for IoMT applications that involve large-scale networks and resource-constrained devices. This paper introduces a novel lightweight sharding method tailored to the unique demands of IoMT data sharing. Our approach enhances state bootstrapping efficiency and reduces operational overhead by utilizing a dual-chain structure comprising a main chain and a snapshot chain. The snapshot chain periodically records key blockchain states, allowing nodes to synchronize more efficiently. This mechanism is critical in reducing the time and resources needed for new nodes to join the network or existing nodes to recover from outages. Additionally, a block state pruning technique is implemented, significantly minimizing storage requirements and lowering transaction execution overhead during initialization and reconfiguration processes. This is crucial given the substantial data volumes inherent in IoMT ecosystems. By adopting an optimistic sharding strategy, our model allows nodes to swiftly join the snapshot shard, while full shards retain the complete ledger history to ensure comprehensive transaction verification. Extensive evaluations across diverse shard configurations demonstrate that this method significantly outperforms existing baseline models. It provides a comprehensive solution for IoMT blockchain applications, striking an optimal balance between security, scalability, and operational efficiency.展开更多
The efficiency of tunnel excavation,rock strength,stability of surrounding rock,and underground engineering disasters are closely related to lithology.Accurately identifying lithology is a necessary prerequisite for i...The efficiency of tunnel excavation,rock strength,stability of surrounding rock,and underground engineering disasters are closely related to lithology.Accurately identifying lithology is a necessary prerequisite for intelligent,safe,and efficient tunnel construction.The design of conventional recognition models heavily relies on experience and extensive calculations.To develop a model suitable for deployment on construction sites and capable of accurate lithology identification,a fast search method for lithology identification models is proposed.This method integrates geological knowledge,apparent feature extraction techniques,and search algorithms.An efficient feature extraction super network using multi-scale geological features of rock surface is constructed,a model evaluation method that comprehensively considers accuracy and latency is developed,and differential evolution algorithm is used to search for the optimal model parameters.Experiments demonstrate that the proposed method enables the model to evolve faster and more accurately,and eventually a model(LithoNet)suitable for lithological classification is found.It only takes 2.10 ms to infer an image of 224×224,which is 57.25%faster than MobileNet v3 and 62.83%faster than ShuffleNet V2.The F1-score of LithoNet is 0.9874,surpassing classical models such as EfficientNetV2-S.LithoNet can be easily deployed on portable devices,effectively promoting the intelligence and accuracy of lithology identification at engineering sites.展开更多
(Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4))_(100−x)Al_(x)(x=0,0.1,0.2,0.3,0.4 at.%)lightweight high-entropy alloys with different contents of Al were prepared via vacuum non-consumable arc melting method.Effects of adding varying...(Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4))_(100−x)Al_(x)(x=0,0.1,0.2,0.3,0.4 at.%)lightweight high-entropy alloys with different contents of Al were prepared via vacuum non-consumable arc melting method.Effects of adding varying Al contents on phase constitution,microstructure characteristics and mechanical properties of the lightweight alloys were studied.Results show that Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4)alloy is composed of body-centered cubic(BCC)phase and C15 Laves phase,while(Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4))_(100−x)Al_(x)lightweight high-entropy alloys by addition of Al are composed of BCC phase and C14 Laves phase.Addition of Al into Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4)lightweight high-entropy alloy can transform C15 Laves phase to C14 Laves phase.With further addition of Al,BCC phase of alloys is significantly refined,and the volume fraction of C14 Laves phase is raised obviously.Meanwhile,the dimension of BCC phase in the alloy by addition of 0.3 at.%Al is the most refined and that of Laves phase is also obviously refined.Adding Al to Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4)alloy can not only reduce the density of(Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4))_(100−x)Al_(x)alloy,but also improve strength of(Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4))_(100−x)Al_(x)alloy.As Al content increased from 0 to 0.4 at.%,the density of the alloy decreased from 6.22±0.875 to 5.79±0.679 g cm^(−3).Moreover,compressive strength of the alloy by 0.3 at.%Al addition is the highest to 1996.9 MPa,while fracture strain of the alloy is 16.82%.Strength improvement of alloys mainly results from microstructure refinement and precipitation of C14 Laves by Al addition into Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4)lightweight high-entropy alloy.展开更多
Traditional Internet of Things(IoT)architectures that rely on centralized servers for data management and decision-making are vulnerable to security threats and privacy leakage.To address this issue,blockchain has bee...Traditional Internet of Things(IoT)architectures that rely on centralized servers for data management and decision-making are vulnerable to security threats and privacy leakage.To address this issue,blockchain has been advocated for decentralized data management in a tamper-resistance,traceable,and transparent manner.However,a major issue that hinders the integration of blockchain and IoT lies in that,it is rather challenging for resource-constrained IoT devices to perform computation-intensive blockchain consensuses such as Proof-of-Work(PoW).Furthermore,the incentive mechanism of PoW pushes lightweight IoT nodes to aggregate their computing power to increase the possibility of successful block generation.Nevertheless,this eventually leads to the formation of computing power alliances,and significantly compromises the decentralization and security of BlockChain-aided IoT(BC-IoT)networks.To cope with these issues,we propose a lightweight consensus protocol for BC-IoT,called Proof-of-Trusted-Work(PoTW).The goal of the proposed consensus is to disincentivize the centralization of computing power and encourage the independent participation of lightweight IoT nodes in blockchain consensus.First,we put forth an on-chain reputation evaluation rule and a reputation chain for PoTW to enable the verifiability and traceability of nodes’reputations based on their contributions of computing power to the blockchain consensus,and we incorporate the multi-level block generation difficulty as a rewards for nodes to accumulate reputations.Second,we model the block generation process of PoTW and analyze the block throughput using the continuous time Markov chain.Additionally,we define and optimize the relative throughput gain to quantify and maximize the capability of PoTW that suppresses the computing power centralization(i.e.,centralization suppression).Furthermore,we investigate the impact of the computing power of the computing power alliance and the levels of block generation difficulty on the centralization suppression capability of PoTW.Finally,simulation results demonstrate the consistency of the analytical results in terms of block throughput.In particular,the results show that PoTW effectively reduces the block generation proportion of the computing power alliance compared with PoW,while simultaneously improving that of individual lightweight nodes.This indicates that PoTW is capable of suppressing the centralization of computing power to a certain degree.Moreover,as the levels of block generation difficulty in PoTW increase,its centralization suppression capability strengthens.展开更多
This study presents the development of a Magnesium Alloy Seat Frame(MASF),supported by case studies from automotive original equipment manufacturers.The process covers integrated design,simulation,manufacturing,and te...This study presents the development of a Magnesium Alloy Seat Frame(MASF),supported by case studies from automotive original equipment manufacturers.The process covers integrated design,simulation,manufacturing,and testing,aiming to boost industry confidence in Mg alloy applications.A novel structural design is developed that integrates the headrest with the backrest,achieving a balance between lightweight performance and safety.Structural optimization is guided by stress–strain simulations under diverse conditions within a complete forward development process.Casting simulations are conducted to analyze process characteristics,resulting in a verified MASF yield rate exceeding 90%.The final 9.88 kg MASF represents a 24.6%(3.23 kg)weight reduction versus a steel seat.This research contributes to advancements in defect control technology for large die casting magnesium alloy parts and has broad implications for their application in automotive manufacturing.展开更多
This paper proposes SW-YOLO(StarNet Weighted-Conv YOLO),a lightweight human pose estimation network for edge devices.Current mainstream pose estimation algorithms are computationally inefficient and have poor feature ...This paper proposes SW-YOLO(StarNet Weighted-Conv YOLO),a lightweight human pose estimation network for edge devices.Current mainstream pose estimation algorithms are computationally inefficient and have poor feature capture capabilities for complex poses and occlusion scenarios.This work introduces a lightweight backbone architecture that integrates WConv(Weighted Convolution)and StarNet modules to address these issues.Leveraging StarNet’s superior capabilities in multi-level feature fusion and long-range dependency modeling,this architecture enhances the model’s spatial perception of human joint structures and contextual information integration.These improvements significantly enhance robustness in complex scenarios involving occlusion and deformation.Additionally,the introduction of WConv convolution operations,based on weight recalibration and receptive field optimization,dynamically adjusts feature importance during convolution.This reduces redundant computations while maintaining or enhancing feature representation capabilities at an extremely low computational cost.Consequently,SW-YOLO substantially reduces model complexity and inference latency while preserving high accuracy,significantly outperforming existing lightweight networks.展开更多
Colorectal cancer is the most common cancer with a second mortality rate.Polyp lesion is a precursor symptom of colorectal cancer.Detection and removal of polyps can effectively reduce the mortality of patients in the...Colorectal cancer is the most common cancer with a second mortality rate.Polyp lesion is a precursor symptom of colorectal cancer.Detection and removal of polyps can effectively reduce the mortality of patients in the early period.However,mass images will be generated during an endoscopy,which will greatly increase the workload of doctors,and long-term mechanical screening of endoscopy images will also lead to a high misdiagnosis rate.Aiming at the problem that computer-aided diagnosis models deeply depend on the computational power in the polyp detection task,we propose a lightweight model,coordinate attention-YOLOv5-Lite-Prune,based on the YOLOv5 algorithm,which is different from state-of-the-art methods proposed by the existing research that applied object detection models or their variants directly to prediction task without any lightweight processing,such as faster region-based convolutional neural networks,YOLOv3,YOLOv4,and single shot multibox detector.The innovations of our model are as follows:First,the lightweight EfficientNetLite network is introduced as the new feature extraction network.Second,the depthwise separable convolution and its improved modules with different attention mechanisms are used to replace the standard convolution in the detection head structure.Then,theα-intersection over union loss function is applied to improve the precision and convergence speed of the model.Finally,the model size is compressed with a pruning algorithm.Our model effectively reduces parameter amount and computational complexity without significant accuracy loss.Therefore,the model can be successfully deployed on the embedded deep learning platform,and detect polyps with a speed above 30 frames per second,which means the model gets rid of the limitation that deep learning models must rely on high-performance servers.展开更多
Lightweight deep learning models are increasingly required in resource-constrained environments such as mobile devices and the Internet of Medical Things(IoMT).Multi-head convolution with channel attention can facilit...Lightweight deep learning models are increasingly required in resource-constrained environments such as mobile devices and the Internet of Medical Things(IoMT).Multi-head convolution with channel attention can facilitate learning activations relevant to different kernel sizes within a multi-head convolutional layer.Therefore,this study investigates the capability of novel lightweight models incorporating residual multi-head convolution with channel attention(ResMHCNN)blocks to classify medical images.We introduced three novel lightweight deep learning models(BT-Net,LCC-Net,and BC-Net)utilizing the ResMHCNN block as their backbone.These models were crossvalidated and tested on three publicly available medical image datasets:a brain tumor dataset from Figshare consisting of T1-weighted magnetic resonance imaging slices of meningioma,glioma,and pituitary tumors;the LC25000 dataset,which includes microscopic images of lung and colon cancers;and the BreaKHis dataset,containing benign and malignant breast microscopic images.The lightweight models achieved accuracies of 96.9%for 3-class brain tumor classification using BT-Net,and 99.7%for 5-class lung and colon cancer classification using LCC-Net.For 2-class breast cancer classification,BC-Net achieved an accuracy of 96.7%.The parameter counts for the proposed lightweight models—LCC-Net,BC-Net,and BT-Net—are 0.528,0.226,and 1.154 million,respectively.The presented lightweight models,featuring ResMHCNN blocks,may be effectively employed for accurate medical image classification.In the future,these models might be tested for viability in resource-constrained systems such as mobile devices and IoMT platforms.展开更多
基金supported by the National Natural Science Foundation of China(Nos.51925103,52271149,52171159)the Innovation Program of Shanghai Municipal Education Commission(No.2021-01-07-00-09-E00114)+5 种基金the Natural Science Foundation of Shanghai(22ZR1422500)the Innovation Program of Shanghai Science and Technology(No.23520760700)the Aviation Foundation(No.2023Z0530S6004)the Fund of the State Key Laboratory of Solidification Processing in NWPU(No.SKLSP202221)the financial support from Program 173(No.2020-JCIQ-ZD-186-01)the Space Utilization System of China Manned Space Engineering(No.KJZ-YY-NCL08).
文摘Lightweight high/medium-entropy alloys(H/MEAs)possess attractive properties such as high strength-to-weight ratios,however,their limited room-temperature tensile ductility hinders their widespread engi-neering implementation,for instance in aerospace structural components.This work achieved a transfor-mative improvement of room-temperature tensile ductility in Ti-V-Zr-Nb MEAs with densities of 5.4-6.5 g/cm3,via ingenious composition modulation.Through the systematic co-adjustment of Ti and V contents,an intrinsic ductility mechanism was unveiled,manifested by a transition from predominant intergranular brittle fracture to pervasive ductile dimpled rupture.Notably,the modulated deformation mechanisms evolved from solitary slip toward collaborative multiple slip modes,without significantly compromising strength.Compared to equimolar Ti-V-Zr-Nb,a(Ti1.5V)3ZrNb composition demonstrated an impressive 360%improvement in elongation while sustaining a high yield strength of around 800 MPa.Increasing Ti and V not only purified the grain boundaries by reducing detrimental phases,but also tai-lored the deformation dislocation configurations.These insights expanded the applicability of lightweight HEAs to areas demanding combined high strength and ductility.
基金financially supported by China Aeronautical Science Foundation (No.2023Z0530Q9002)the Program for Chongqing Talents,China (No.cstc2024ycjh-bgzxm0066)。
文摘The microstructural evolution,phase stability,and mechanical properties of Al-Li-Mg-Ti-M(M=Zn,Zr,V)lightweight high-entropy alloys(LW-HEAs)were investigated.The LW-HEAs with three components,Al_(20)Li_(20)Mg_(10)-Ti_(40)Zn_(10)(#Zn),Al_(20)Li_(20)Mg_(10)Ti_(30)Zr_(20)(#Zr),and Al_(20)Li_(20)Mg_(10)Ti_(30)V_(20)(#V),were designed according to the thermo-dynamic design criteria of HEA,and prepared via a combination process of mechanical alloying and cold-press sintering.The effects of alloy composition and sintering temperature on the microstructure and mechanical properties of the LW-HEAs were studied.The results show that the as-milled Al-Li-Mg-Ti-M(M=Zn,Zr,V)LW-HEAs form a simple structure with HCP-type solid solution as the primary phase,a dual-HCP type solid solution phase,and a BCC phase,respectively.After cold-press sintering,the#Zn and#V alloys undergo obvious phase transformation;while the#Zr alloy with dual-HCP phases exhibits the best phase stability during heat treatment.The#V-750°C alloy demonstrates the maximum hardness and specific strength of HV 595.2 and 625 MPa∙cm3/g,respectively,under the combined effect of solid solution strengthening of BCC phase and precipitation strengthening ofβ-AlTi_(3).Moreover,the#Zr-650°C,#Zr-750°C,and#Zn-650°C alloys are expected to have excellent plasticity.
文摘Tomato plant diseases often first manifest on the leaves,making the detection of tomato leaf diseases particularly crucial for the tomato cultivation industry.However,conventional deep learning models face challenges such as large model sizes and slow detection speeds when deployed on resource-constrained platforms and agricultural machinery.This paper proposes a lightweight model for detecting tomato leaf diseases,named LT-YOLO,based on the YOLOv8n architecture.First,we enhance the C2f module into a RepViT Block(RVB)with decoupled token and channel mixers to reduce the cost of feature extraction.Next,we incorporate a novel Efficient Multi-Scale Attention(EMA)mechanism in the deeper layers of the backbone to improve detection of critical disease features.Additionally,we design a lightweight detection head,LT-Detect,using Partial Convolution(PConv)to significantly reduce the classification and localization costs during detection.Finally,we introduce a Receptive Field Block(RFB)in the shallow layers of the backbone to expand the model’s receptive field,enabling effective detection of diseases at various scales.The improved model reduces the number of parameters by 43%and the computational load by 50%.Additionally,it achieves a mean Average Precision(mAP)of 90.9%on a publicly available dataset containing 3641 images of tomato leaf diseases,with only a 0.7%decrease compared to the baseline model.This demonstrates that the model maintains excellent accuracy while being lightweight,making it suitable for rapid detection of tomato leaf diseases.
基金supported by the National Natural Science Foundation of China(22379021 and 22479021)。
文摘As batteries become increasingly essential for energy storage technologies,battery prognosis,and diagnosis remain central to ensure reliable operation and effective management,as well as to aid the in-depth investigation of degradation mechanisms.However,dynamic operating conditions,cell-to-cell inconsistencies,and limited availability of labeled data have posed significant challenges to accurate and robust prognosis and diagnosis.Herein,we introduce a time-series-decomposition-based ensembled lightweight learning model(TELL-Me),which employs a synergistic dual-module framework to facilitate accurate and reliable forecasting.The feature module formulates features with physical implications and sheds light on battery aging mechanisms,while the gradient module monitors capacity degradation rates and captures aging trend.TELL-Me achieves high accuracy in end-of-life prediction using minimal historical data from a single battery without requiring offline training dataset,and demonstrates impressive generality and robustness across various operating conditions and battery types.Additionally,by correlating feature contributions with degradation mechanisms across different datasets,TELL-Me is endowed with the diagnostic ability that not only enhances prediction reliability but also provides critical insights into the design and optimization of next-generation batteries.
文摘The Internet of Things(IoT)has gained substantial attention in both academic research and real-world applications.The proliferation of interconnected devices across various domains promises to deliver intelligent and advanced services.However,this rapid expansion also heightens the vulnerability of the IoT ecosystem to security threats.Consequently,innovative solutions capable of effectively mitigating risks while accommodating the unique constraints of IoT environments are urgently needed.Recently,the convergence of Blockchain technology and IoT has introduced a decentralized and robust framework for securing data and interactions,commonly referred to as the Internet of Blockchained Things(IoBT).Extensive research efforts have been devoted to adapting Blockchain technology to meet the specific requirements of IoT deployments.Within this context,consensus algorithms play a critical role in assessing the feasibility of integrating Blockchain into IoT ecosystems.The adoption of efficient and lightweight consensus mechanisms for block validation has become increasingly essential.This paper presents a comprehensive examination of lightweight,constraint-aware consensus algorithms tailored for IoBT.The study categorizes these consensus mechanisms based on their core operations,the security of the block validation process,the incorporation of AI techniques,and the specific applications they are designed to support.
基金supported by the National Natural Science Foundation of China(Nos.52171166 and U20A20231)the Natural Science Foundation of Hunan Province,China(Nos.2024JJ2060 and 2024JJ5406)+1 种基金the Key Laboratory of Materials in Dynamic Extremes of Sichuan Province(No.2023SCKT1102)the Postgraduate Scientific Research Innovation Project of National University of Defense Technology(No.XJJC2024065).
文摘Introducing B2 ordering can effectively improve the mechanical properties of lightweight refractory high-entropy alloys(LRHEAs).However,(Zr,Al)-enriched B2 precipitates generally reduce the ductility because their ordering characteristic is destroyed after dislocation shearing.Meanwhile,the local chemical order(LCO)cannot provide an adequate strengthening effect due to its small size.
文摘In recent years,with the development of synthetic aperture radar(SAR)technology and the widespread application of deep learning,lightweight detection of SAR images has emerged as a research direction.The ultimate goal is to reduce computational and storage requirements while ensuring detection accuracy and reliability,making it an ideal choice for achieving rapid response and efficient processing.In this regard,a lightweight SAR ship target detection algorithm based on YOLOv8 was proposed in this study.Firstly,the C2f-Sc module was designed by fusing the C2f in the backbone network with the ScConv to reduce spatial redundancy and channel redundancy between features in convolutional neural networks.At the same time,the Ghost module was introduced into the neck network to effectively reduce model parameters and computational complexity.A relatively lightweight EMA attention mechanism was added to the neck network to promote the effective fusion of features at different levels.Experimental results showed that the Parameters and GFLOPs of the improved model are reduced by 8.5%and 7.0%when mAP@0.5 and mAP@0.5:0.95 are increased by 0.7%and 1.8%,respectively.It makes the model lightweight and improves the detection accuracy,which has certain application value.
基金supported by the HFIPS Director’s Foundation(YZJJ202207-TS),the National Natural Science Foundation of China(82371931)the Natural Science Foundation of Anhui Province(2008085MC69)+3 种基金the Natural Science Foundation of Hefei City(2021033)the General Scientific Research Project of Anhui Provincial Health Commission(AHWJ2021b150)the Collaborative Innovation Program of Hefei Science Center,CAS(2021HSC-CIP013)the Anhui Province Key Research and Development Project(202204295107020004).
文摘Semantic segmentation of eye images is a complex task with important applications in human–computer interaction,cognitive science,and neuroscience.Achieving real-time,accurate,and robust segmentation algorithms is crucial for computationally limited portable devices such as augmented reality and virtual reality.With the rapid advancements in deep learning,many network models have been developed specifically for eye image segmentation.Some methods divide the segmentation process into multiple stages to achieve model parameter miniaturization while enhancing output through post processing techniques to improve segmentation accuracy.These approaches significantly increase the inference time.Other networks adopt more complex encoding and decoding modules to achieve end-to-end output,which requires substantial computation.Therefore,balancing the model’s size,accuracy,and computational complexity is essential.To address these challenges,we propose a lightweight asymmetric UNet architecture and a projection loss function.We utilize ResNet-3 layer blocks to enhance feature extraction efficiency in the encoding stage.In the decoding stage,we employ regular convolutions and skip connections to upscale the feature maps from the latent space to the original image size,balancing the model size and segmentation accuracy.In addition,we leverage the geometric features of the eye region and design a projection loss function to further improve the segmentation accuracy without adding any additional inference computational cost.We validate our approach on the OpenEDS2019 dataset for virtual reality and achieve state-of-the-art performance with 95.33%mean intersection over union(mIoU).Our model has only 0.63M parameters and 350 FPS,which are 68%and 200%of the state-of-the-art model RITNet,respectively.
基金supported by the Shanghai Sailing Program,China (No.20YF1447600)the Research Start-Up Project of Shanghai Institute of Technology (No.YJ2021-60)+1 种基金the Collaborative Innovation Project of Shanghai Institute of Technology (No.XTCX2020-12)the Science and Technology Talent Development Fund for Young and Middle-Aged Teachers at Shanghai Institute of Technology (No.ZQ2022-6)。
文摘There is a problem of real-time detection difficulty in road surface damage detection. This paper proposes an improved lightweight model based on you only look once version 5(YOLOv5). Firstly, this paper fully utilized the convolutional neural network(CNN) + ghosting bottleneck(G_bneck) architecture to reduce redundant feature maps. Afterwards, we upgraded the original upsampling algorithm to content-aware reassembly of features(CARAFE) and increased the receptive field. Finally, we replaced the spatial pyramid pooling fast(SPPF) module with the basic receptive field block(Basic RFB) pooling module and added dilated convolution. After comparative experiments, we can see that the number of parameters and model size of the improved algorithm in this paper have been reduced by nearly half compared to the YOLOv5s. The frame rate per second(FPS) has been increased by 3.25 times. The mean average precision(m AP@0.5: 0.95) has increased by 8%—17% compared to other lightweight algorithms.
基金supported by the National Natural Science Foundation of China(No.62103298)the Natural Science Foundation of Hebei Province(No.F2018209289)。
文摘Aiming at the problems of low detection efficiency and difficult positioning of traditional steel surface defect detection methods,a lightweight steel surface defect detection model based on you only look once version 7(YOLOv7)is proposed.First,a cascading style sheets(CSS)block module is proposed,which uses more lightweight operations to obtain redundant information in the feature map,reduces the amount of computation,and effectively improves the detection speed.Secondly,the improved spatial pyramid pooling with cross stage partial convolutions(SPPCSPC)structure is adopted to ensure that the model can also pay attention to the defect location information while predicting the defect category information,obtain richer defect features.In addition,the convolution operation in the original model is simplified,which significantly reduces the size of the model and helps to improve the detection speed.Finally,using efficient intersection over union(EIOU)loss to focus on high-quality anchors,speed up convergence and improve positioning accuracy.Experiments were carried out on the Northeastern University-defect(NEU-DET)steel surface defect dataset.Compared with the original YOLOv7 model,the number of parameters of this model was reduced by 40%,the frames per second(FPS)reached 112,and the average accuracy reached 79.1%,the detection accuracy and speed have been improved,which can meet the needs of steel surface defect detection.
基金supported by the National Natural Science Foundation of China(Nos.62373215,62373219 and 62073193)the Natural Science Foundation of Shandong Province(No.ZR2023MF100)+1 种基金the Key Projects of the Ministry of Industry and Information Technology(No.TC220H057-2022)the Independently Developed Instrument Funds of Shandong University(No.zy20240201)。
文摘Current you only look once(YOLO)-based algorithm model is facing the challenge of overwhelming parameters and calculation complexity under the printed circuit board(PCB)defect detection application scenario.In order to solve this problem,we propose a new method,which combined the lightweight network mobile vision transformer(Mobile Vi T)with the convolutional block attention module(CBAM)mechanism and the new regression loss function.This method needed less computation resources,making it more suitable for embedded edge detection devices.Meanwhile,the new loss function improved the positioning accuracy of the bounding box and enhanced the robustness of the model.In addition,experiments on public datasets demonstrate that the improved model achieves an average accuracy of 87.9%across six typical defect detection tasks,while reducing computational costs by nearly 90%.It significantly reduces the model's computational requirements while maintaining accuracy,ensuring reliable performance for edge deployment.
基金supported by the National Natural Science Foundation of China(62272207)the Key Project of Natural Science Foundation of Jiangxi Province(20224ACB202009)+1 种基金the Science and Technology Project of theDepartment of Education of Jiangxi Province(GJJ2200925)the Jiangxi Provincial Health Commission Science and Technology Plan(202311147).
文摘As the Internet of Medical Things (IoMT) continues to expand, smart health-monitoring devices generate vast amounts of valuable data while simultaneously raising critical security and privacy challenges. Blockchain technology presents a promising avenue to address these concerns due to its inherent decentralization and security features. However, scalability remains a persistent hurdle, particularly for IoMT applications that involve large-scale networks and resource-constrained devices. This paper introduces a novel lightweight sharding method tailored to the unique demands of IoMT data sharing. Our approach enhances state bootstrapping efficiency and reduces operational overhead by utilizing a dual-chain structure comprising a main chain and a snapshot chain. The snapshot chain periodically records key blockchain states, allowing nodes to synchronize more efficiently. This mechanism is critical in reducing the time and resources needed for new nodes to join the network or existing nodes to recover from outages. Additionally, a block state pruning technique is implemented, significantly minimizing storage requirements and lowering transaction execution overhead during initialization and reconfiguration processes. This is crucial given the substantial data volumes inherent in IoMT ecosystems. By adopting an optimistic sharding strategy, our model allows nodes to swiftly join the snapshot shard, while full shards retain the complete ledger history to ensure comprehensive transaction verification. Extensive evaluations across diverse shard configurations demonstrate that this method significantly outperforms existing baseline models. It provides a comprehensive solution for IoMT blockchain applications, striking an optimal balance between security, scalability, and operational efficiency.
基金financial support from the National Natural Science Foundation of China(Grant Nos.52279103 and 52379103)the Natural Science Foundation of Shandong Province(Grant No.ZR2023YQ049).
文摘The efficiency of tunnel excavation,rock strength,stability of surrounding rock,and underground engineering disasters are closely related to lithology.Accurately identifying lithology is a necessary prerequisite for intelligent,safe,and efficient tunnel construction.The design of conventional recognition models heavily relies on experience and extensive calculations.To develop a model suitable for deployment on construction sites and capable of accurate lithology identification,a fast search method for lithology identification models is proposed.This method integrates geological knowledge,apparent feature extraction techniques,and search algorithms.An efficient feature extraction super network using multi-scale geological features of rock surface is constructed,a model evaluation method that comprehensively considers accuracy and latency is developed,and differential evolution algorithm is used to search for the optimal model parameters.Experiments demonstrate that the proposed method enables the model to evolve faster and more accurately,and eventually a model(LithoNet)suitable for lithological classification is found.It only takes 2.10 ms to infer an image of 224×224,which is 57.25%faster than MobileNet v3 and 62.83%faster than ShuffleNet V2.The F1-score of LithoNet is 0.9874,surpassing classical models such as EfficientNetV2-S.LithoNet can be easily deployed on portable devices,effectively promoting the intelligence and accuracy of lithology identification at engineering sites.
基金supported by National Natural Science Foundation of China(Grant No.52001114)Program for Science and Technology Innovation Talents in Universities of Henan Province(No.23HASTIT022 and 2021GGJS064)Scientific Research Fund of State Key Laboratory of Materials Processing and Die and Mould Technology(Grant No.P2023-005).
文摘(Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4))_(100−x)Al_(x)(x=0,0.1,0.2,0.3,0.4 at.%)lightweight high-entropy alloys with different contents of Al were prepared via vacuum non-consumable arc melting method.Effects of adding varying Al contents on phase constitution,microstructure characteristics and mechanical properties of the lightweight alloys were studied.Results show that Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4)alloy is composed of body-centered cubic(BCC)phase and C15 Laves phase,while(Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4))_(100−x)Al_(x)lightweight high-entropy alloys by addition of Al are composed of BCC phase and C14 Laves phase.Addition of Al into Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4)lightweight high-entropy alloy can transform C15 Laves phase to C14 Laves phase.With further addition of Al,BCC phase of alloys is significantly refined,and the volume fraction of C14 Laves phase is raised obviously.Meanwhile,the dimension of BCC phase in the alloy by addition of 0.3 at.%Al is the most refined and that of Laves phase is also obviously refined.Adding Al to Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4)alloy can not only reduce the density of(Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4))_(100−x)Al_(x)alloy,but also improve strength of(Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4))_(100−x)Al_(x)alloy.As Al content increased from 0 to 0.4 at.%,the density of the alloy decreased from 6.22±0.875 to 5.79±0.679 g cm^(−3).Moreover,compressive strength of the alloy by 0.3 at.%Al addition is the highest to 1996.9 MPa,while fracture strain of the alloy is 16.82%.Strength improvement of alloys mainly results from microstructure refinement and precipitation of C14 Laves by Al addition into Ti_(8)Zr_(6)Nb_(4)V_(5)Cr_(4)lightweight high-entropy alloy.
基金supported in part by National Key R&D Program of China(Grant No.2021YFB1714100)in part by the National Natural Science Foundation of China(NSFC)under Grant 62371239+5 种基金in part by the the Program of Science and Technology Cooperation of Nanjing with International/Hong Kong,Macao and Taiwan(Grant No.202401019)in part by the Guangdong Basic and Applied Basic Research Foundation(Grant No.2024A1515012407)in part by the the Research Center for FinTech and Digital-Intelligent Management at Shenzhen University,in part by the National Natural Science Foundation of China under Grant 62271192in part by the Equipment Pre-Research Joint Research Program of Ministry of Education under Grant 8091B032129in part by the Major Science and Technology Projects of Longmen Laboratory under Grant 231100220300 and 231100220200in part by the Central Plains Leading Talent in Scientific and Technological Innovation Program under Grant 244200510048.
文摘Traditional Internet of Things(IoT)architectures that rely on centralized servers for data management and decision-making are vulnerable to security threats and privacy leakage.To address this issue,blockchain has been advocated for decentralized data management in a tamper-resistance,traceable,and transparent manner.However,a major issue that hinders the integration of blockchain and IoT lies in that,it is rather challenging for resource-constrained IoT devices to perform computation-intensive blockchain consensuses such as Proof-of-Work(PoW).Furthermore,the incentive mechanism of PoW pushes lightweight IoT nodes to aggregate their computing power to increase the possibility of successful block generation.Nevertheless,this eventually leads to the formation of computing power alliances,and significantly compromises the decentralization and security of BlockChain-aided IoT(BC-IoT)networks.To cope with these issues,we propose a lightweight consensus protocol for BC-IoT,called Proof-of-Trusted-Work(PoTW).The goal of the proposed consensus is to disincentivize the centralization of computing power and encourage the independent participation of lightweight IoT nodes in blockchain consensus.First,we put forth an on-chain reputation evaluation rule and a reputation chain for PoTW to enable the verifiability and traceability of nodes’reputations based on their contributions of computing power to the blockchain consensus,and we incorporate the multi-level block generation difficulty as a rewards for nodes to accumulate reputations.Second,we model the block generation process of PoTW and analyze the block throughput using the continuous time Markov chain.Additionally,we define and optimize the relative throughput gain to quantify and maximize the capability of PoTW that suppresses the computing power centralization(i.e.,centralization suppression).Furthermore,we investigate the impact of the computing power of the computing power alliance and the levels of block generation difficulty on the centralization suppression capability of PoTW.Finally,simulation results demonstrate the consistency of the analytical results in terms of block throughput.In particular,the results show that PoTW effectively reduces the block generation proportion of the computing power alliance compared with PoW,while simultaneously improving that of individual lightweight nodes.This indicates that PoTW is capable of suppressing the centralization of computing power to a certain degree.Moreover,as the levels of block generation difficulty in PoTW increase,its centralization suppression capability strengthens.
基金supported in part by the project is supported partly by National Key Research and Development Program of China(no.2022YFB2503504)Chongqing Technology Innovation and Application Development Project(no.CSTB2022TIAD-DEX0011)China Scholarship Council.
文摘This study presents the development of a Magnesium Alloy Seat Frame(MASF),supported by case studies from automotive original equipment manufacturers.The process covers integrated design,simulation,manufacturing,and testing,aiming to boost industry confidence in Mg alloy applications.A novel structural design is developed that integrates the headrest with the backrest,achieving a balance between lightweight performance and safety.Structural optimization is guided by stress–strain simulations under diverse conditions within a complete forward development process.Casting simulations are conducted to analyze process characteristics,resulting in a verified MASF yield rate exceeding 90%.The final 9.88 kg MASF represents a 24.6%(3.23 kg)weight reduction versus a steel seat.This research contributes to advancements in defect control technology for large die casting magnesium alloy parts and has broad implications for their application in automotive manufacturing.
文摘This paper proposes SW-YOLO(StarNet Weighted-Conv YOLO),a lightweight human pose estimation network for edge devices.Current mainstream pose estimation algorithms are computationally inefficient and have poor feature capture capabilities for complex poses and occlusion scenarios.This work introduces a lightweight backbone architecture that integrates WConv(Weighted Convolution)and StarNet modules to address these issues.Leveraging StarNet’s superior capabilities in multi-level feature fusion and long-range dependency modeling,this architecture enhances the model’s spatial perception of human joint structures and contextual information integration.These improvements significantly enhance robustness in complex scenarios involving occlusion and deformation.Additionally,the introduction of WConv convolution operations,based on weight recalibration and receptive field optimization,dynamically adjusts feature importance during convolution.This reduces redundant computations while maintaining or enhancing feature representation capabilities at an extremely low computational cost.Consequently,SW-YOLO substantially reduces model complexity and inference latency while preserving high accuracy,significantly outperforming existing lightweight networks.
基金the National Natural Science Foundation of China(Nos.81971767,62103263 and 62103267)the Shanghai Science and Technology Commission(Nos.19142203800,19441913800 and 19441910600)。
文摘Colorectal cancer is the most common cancer with a second mortality rate.Polyp lesion is a precursor symptom of colorectal cancer.Detection and removal of polyps can effectively reduce the mortality of patients in the early period.However,mass images will be generated during an endoscopy,which will greatly increase the workload of doctors,and long-term mechanical screening of endoscopy images will also lead to a high misdiagnosis rate.Aiming at the problem that computer-aided diagnosis models deeply depend on the computational power in the polyp detection task,we propose a lightweight model,coordinate attention-YOLOv5-Lite-Prune,based on the YOLOv5 algorithm,which is different from state-of-the-art methods proposed by the existing research that applied object detection models or their variants directly to prediction task without any lightweight processing,such as faster region-based convolutional neural networks,YOLOv3,YOLOv4,and single shot multibox detector.The innovations of our model are as follows:First,the lightweight EfficientNetLite network is introduced as the new feature extraction network.Second,the depthwise separable convolution and its improved modules with different attention mechanisms are used to replace the standard convolution in the detection head structure.Then,theα-intersection over union loss function is applied to improve the precision and convergence speed of the model.Finally,the model size is compressed with a pruning algorithm.Our model effectively reduces parameter amount and computational complexity without significant accuracy loss.Therefore,the model can be successfully deployed on the embedded deep learning platform,and detect polyps with a speed above 30 frames per second,which means the model gets rid of the limitation that deep learning models must rely on high-performance servers.
基金supported by the Institute of Information&Communications Technology Planning&Evaluation(IITP)-Innovative Human Resource Development for Local Intellectualization program grant funded by the Korea government(MSIT)(IITP-2025-RS-2023-00259678)by INHA UNIVERSITY Research Grant.
文摘Lightweight deep learning models are increasingly required in resource-constrained environments such as mobile devices and the Internet of Medical Things(IoMT).Multi-head convolution with channel attention can facilitate learning activations relevant to different kernel sizes within a multi-head convolutional layer.Therefore,this study investigates the capability of novel lightweight models incorporating residual multi-head convolution with channel attention(ResMHCNN)blocks to classify medical images.We introduced three novel lightweight deep learning models(BT-Net,LCC-Net,and BC-Net)utilizing the ResMHCNN block as their backbone.These models were crossvalidated and tested on three publicly available medical image datasets:a brain tumor dataset from Figshare consisting of T1-weighted magnetic resonance imaging slices of meningioma,glioma,and pituitary tumors;the LC25000 dataset,which includes microscopic images of lung and colon cancers;and the BreaKHis dataset,containing benign and malignant breast microscopic images.The lightweight models achieved accuracies of 96.9%for 3-class brain tumor classification using BT-Net,and 99.7%for 5-class lung and colon cancer classification using LCC-Net.For 2-class breast cancer classification,BC-Net achieved an accuracy of 96.7%.The parameter counts for the proposed lightweight models—LCC-Net,BC-Net,and BT-Net—are 0.528,0.226,and 1.154 million,respectively.The presented lightweight models,featuring ResMHCNN blocks,may be effectively employed for accurate medical image classification.In the future,these models might be tested for viability in resource-constrained systems such as mobile devices and IoMT platforms.
