检索结果-南通市图书馆

A Study on Classification and Detection of Small Moths Using CNN Model

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2022年第4期71卷 1987-1998页

作者： sang-hyun lee Department of Computer Engineering Honam UniversityGwangsanguGwangju62399South Korea

Currently,there are many limitations to classify images of small objects.In addition,there are limitations such as error detection due to external factors,and there is also a disadvantage that it is difficult to accurately distinguish between various objects.This paper uses a convolutional neural network(CNN)algorithm to recognize and classify object images of very small moths and obtain precise data images.A convolution neural network algorithm is used for image data classification,and the classified image is transformed into image data to learn the topological structure of the image.To improve the accuracy of the image classification and reduce the loss rate,a parameter for finding a fast-optimal point of image classification is set by a convolutional neural network and a pixel image as a preprocessor.As a result of this study,we applied a convolution neural network algorithm to classify the images of very small moths by capturing precise images of the moths.Experimental results showed that the accuracy of classification of very small moths was more than 90%.

关键词： Convolution neural network rectified linear unit activation function pooling feature map

A Study on Cascade R-CNN-Based Dangerous Goods Detection Using X-Ray Image

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2022年第11期73卷 4245-4260页

作者： sang-hyun lee Department of Computer Engineering Honam UniversityGwangsanguGwangju 62399Korea

X-ray inspection equipment is divided into small baggage inspection equipment and large cargo inspection equipment.In the case of inspection using X-ray scanning equipment,it is possible to identify the contents of goods,unauthorized transport,or hidden goods in real-time by-passing cargo through X-rays without opening it.In this paper,we propose a system for detecting dangerous objects in X-ray images using the Cascade Region-based Convolutional Neural Network(Cascade R-CNN)model,and the data used for learning consists of dangerous goods,storage media,firearms,and knives.In addition,to minimize the overfitting problem caused by the lack of data to be used for artificial intelligence(AI)training,data samples are increased by using the CP(copy-paste)algorithm on the existing data.It also solves the data labeling problem by mixing supervised and semi-supervised learning.The four comparative models to be used in this study are Faster Regionbased Convolutional Neural Networks Residual2 Network-101(Faster R-CNN_Res2Net-101)supervised learning,Cascade R-CNN_Res2Net-101_supervised learning,Cascade Region-based Convolutional Neural Networks Composite Backbone Network V2(CBNetV2)Network-101(Cascade R-CNN_CBNetV2Net-101)_supervised learning,and Cascade RCNN_CBNetV2-101_semi-supervised learning which are then compared and evaluated.As a result of comparing the performance of the four models in this paper,in case of Cascade R-CNN_CBNetV2-101_semi-supervised learning,Average Precision(AP)(Intersection over Union(IoU)=0.5):0.7%,AP(IoU=0.75):1.0%than supervised learning,Recall:0.8%higher.

关键词： Cascade R-CNN model faster R-CNN model X-ray screening equipment Res2Net supervised learning semi-supervised learning

Adverse effects of early-life stress:focus on the rodent neuroendocrine system

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Neural Regeneration Research 2024年第2期19卷 336-341页

作者： Seung hyun lee Eui-Man Jung Department of Molecular Biology College of Natural SciencesPusan National UniversityBusanRepublic of Korea

Early-life stress is associated with a high prevalence of mental illnesses such as post-traumatic stress disorders,attention-deficit/hyperactivity disorder,schizophrenia,and anxiety or depressive behavior,which constitute major public health problems.In the early stages of brain development after birth,events such as synaptogenesis,neuron maturation,and glial differentiation occur in a highly orchestrated manner,and external stress can cause adverse long-term effects throughout life.Our body utilizes multifaceted mechanisms,including neuroendocrine and neurotransmitter signaling pathways,to appropriately process external stress.Newborn individuals first exposed to early-life stress deploy neurogenesis as a stress-defense mechanism;however,in adulthood,early-life stress induces apoptosis of mature neurons,activation of immune responses,and reduction of neurotrophic factors,leading to anxiety,depression,and cognitive and memory dysfunction.This process involves the hypothalamus-pituitary-adrenal axis and neurotransmitters secreted by the central nervous system,including norepinephrine,dopamine,and serotonin.The rodent early-life stress model is generally used to experimentally assess the effects of stress during neurodevelopment.This paper reviews the use of the early-life stress model and stress response mechanisms of the body and discusses the experimental results regarding how early-life stress mediates stress-related pathways at a high vulnerability of psychiatric disorder in adulthood.

关键词： early-life stress hypothalamic-pituitary-adrenergic axis maternal separation mental illness neurodevelopmental disorder neuroendocrine system neurotransmitter

Small but big leaps towards neuroglycomics:exploring N-glycome in the brain to advance the understanding of brain development and function

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

Neural Regeneration Research 2024年第3期19卷 489-490页

作者： Boyoung lee hyun Joo An Center for Cognition and Sociality Institute for Basic Science(IBS)DaejeonSouth Korea Asia-Pacific Glycomics Reference Site DaejeonSouth Korea Graduate School of Analytical Science&Technology Chungnam National UniversityDaejeonSouth Korea

Glycosylation is a process that involves the addition of sugar moieties or glycans to different types of molecules,including proteins,lipids,and nucleic acids.Among these,protein glycosylation is one of the most prevalent forms of post-translational modification,playing a crucial role in biological complexity.With more than ten monosaccharides identified within mammalian brain cells and more than 1×1012 possible combinations,the heterogeneity of glycosylation is extensive(Conroy et al.,2021).The diversity of glycans and the complexity of their structures allow for a wide range of protein functions.N-glycans are one of the most abundant forms of glycans and are involved in various cellular functions.N-glycans can be added to proteins at specific sequons,Asn-X-Ser/Thr,and are classified into three main types in mature glycoproteins:high mannose,complex,and hybrid.High mannose N-glycans consist of 5-9 mannose residues linked to a chitobiose core and undergo processing into complex or hybrid forms in the Golgi apparatus(Varki et al.,2017).Complex N-glycans are more diverse and contain various branched structures such as antennae with fucose,galactose,and sialic acid residues.Hybrid N-glycans contain one or more complex branches in conjunction with an oligomannose branch(Fisher and Ungar,2016).Understanding the specific functions of these different types of N-glycans in protein regulation,folding,and function is an active area of research in the life sciences,including glycobiology.

关键词： functions function apparatus

Dual-logic-in-memory implementation with orthogonal polarization of van der Waals ferroelectric heterostructure

维普期刊数据库

同方期刊数据库评论

在线全文

学校读者我要写书评

暂无评论

InfoMat 2024年第2期6卷 107-117页

作者： Jingjie Niu Sumin Jeon Donggyu Kim Sungpyo Baek hyun Ho Yoo Jie Li Ji-sang Park Yoonmyung lee Sungjoo lee SKKU Advanced Institute of Nanotechnology(SAINT) Sungkyunkwan UniversitySuwonRepublic of Korea Department of Nano Science and Technology Sungkyunkwan UniversitySuwonRepublic of Korea Department of Electrical and Computer Engineering Sungkyunkwan UniversitySuwonRepublic of Korea IMEC HeverleeBelgium Department of Nano Engineering Sungkyunkwan UniversitySuwonRepublic of Korea

The rapid advancement of AI-enabled applications has resulted in an increasing need for energy-efficient computing hardware.Logic-in-memory is a promising approach for processing the data stored in memory,wherein fast and efficient computations are possible owing to the parallel execution of reconfigurable logic operations.In this study,a dual-logic-in-memory device,which can simultaneously perform two logic operations in four states,is demonstrated using van der Waals ferroelectric field-effect transistors(vdW FeFETs).The proposed dual-logic-in-memory device,which also acts as a twobit storage device,is a single bidirectional polarization-integrated ferroelectric field-effect transistor(BPI-FeFET).It is fabricated by integrating an in-plane vdW ferroelectric semiconductor SnS and an out-of-plane vdW ferroelectric gate dielectric material—CuInP_(2)S_(6).Four reliable resistance states with excellent endurance and retention characteristics were achieved.The two-bit storage mechanism in a BPI-FeFET was analyzed from two perspectives:carrier density and carrier injection controls,which originated from the out-of-plane polarization of the gate dielectric and in-plane polarization of the semiconductor,respectively.Unlike conventional multilevel FeFETs,the proposed BPIFeFET does not require additional pre-examination or erasing steps to switch from/to an intermediate polarization,enabling direct switching between the four memory states.To utilize the fabricated BPI-FeFET as a dual-logic-inmemory device,two logical operations were selected(XOR and AND),and their parallel execution was demonstrated.Different types of logic operations could be implemented by selecting different initial states,demonstrating various types of functions required for numerous neural network operations.The flexibility and efficiency of the proposed dual-logic-in-memory device appear promising in the realization of next-generation low-power computing systems.

关键词： ferroelectric field-effect transistor in-plane ferroelectricity logic-in-memory out-of-plane ferroelectricity

A Study on Small Pest Detection Based on a CascadeR-CNN-Swin Model

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2022年第9期72卷 6155-6165页

作者： Man-Ting Li sang-hyun lee Department of Computer Engineering Honam UniversityGwangsanguGwangju62399Korea

This study aims to detect and prevent greening disease in citrus trees using a deep neural network.The process of collecting data on citrus greening disease is very difficult because the vector pests are too small.In this paper,since the amount of data collected for deep learning is insufficient,we intend to use the efficient feature extraction function of the neural network based on the Transformer algorithm.We want to use the Cascade Region-based Convolutional Neural Networks(Cascade R-CNN)Swin model,which is a mixture of the transformer model and Cascade R-CNN model to detect greening disease occurring in citrus.In this paper,we try to improve model safety by establishing a linear relationship between samples using Mixup and Cutmix algorithms,which are image processing-based data augmentation techniques.In addition,by using the ImageNet dataset,transfer learning,and stochastic weight averaging(SWA)methods,more accuracy can be obtained.This study compared the Faster Region-based Convolutional Neural Networks Residual Network101(Faster R-CNN ResNet101)model,Cascade Regionbased Convolutional Neural Networks Residual Network101(Cascade RCNN-ResNet101)model,and Cascade R-CNN Swin Model.As a result,the Faster R-CNN ResNet101 model came out as Average Precision(AP)(Intersection over Union(IoU)=0.5):88.2%,AP(IoU=0.75):62.8%,Recall:68.2%,and the Cascade R-CNN ResNet101 model was AP(IoU=0.5):91.5%,AP(IoU=0.75):67.2%,Recall:73.1%.Alternatively,the Cascade R-CNN Swin Model showed AP(IoU=0.5):94.9%,AP(IoU=0.75):79.8%and Recall:76.5%.Thus,the Cascade R-CNN Swin Model showed the best results for detecting citrus greening disease.

关键词： Cascade R-CNN swin model cascade R-CNN resNet101 model faster R-CNN ResNet101 model mixup cutmix

Enhancing Identity Protection in Metaverse-Based Psychological Counseling System

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第1期78卷 617-632页

作者： Jun lee Hanna lee Seong Chan lee hyun Kwon Department of Game Software Hoseo UniversityAsan-si31499Korea Yatav Enter Seoul04799Korea Department of Artificial Intelligence and Data Science Korea Military AcademySeoul01805Korea

Non-face-to-face psychological counseling systems rely on network technologies to anonymize information regard-ing client identity.However,these systems often face challenges concerning voice data leaks and the suboptimal communication of the client’s non-verbal expressions,such as facial cues,to the counselor.This study proposes a metaverse-based psychological counseling system designed to enhance client identity protection while ensuring efficient information delivery to counselors during non-face-to-face counseling.The proposed systemincorporates a voicemodulation function that instantlymodifies/masks the client’s voice to safeguard their identity.Additionally,it employs real-time client facial expression recognition using an ensemble of decision trees to mirror the client’s non-verbal expressions through their avatar in the metaverse environment.The system is adaptable for use on personal computers and smartphones,offering users the flexibility to access metaverse-based psychological counseling across diverse environments.The performance evaluation of the proposed system confirmed that the voice modulation and real-time facial expression replication consistently achieve an average speed of 48.32 frames per second or higher,even when tested on the least powerful smartphone configurations.Moreover,a total of 550 actual psychological counseling sessions were conducted,and the average satisfaction rating reached 4.46 on a 5-point scale.This indicates that clients experienced improved identity protection compared to conventional non-face-to-face metaverse counseling approaches.Additionally,the counselor successfully addressed the challenge of conveying non-verbal cues from clients who typically struggled with non-face-to-face psychological counseling.The proposed systemholds significant potential for applications in interactive discussions and educational activities in the metaverse.

关键词： Metaverse counseling system face tracking identity protection

Depolarization mitigated in ferroelectric Hf_(0.5)Zr_(0.5)O_(2)ultrathin films(<5 nm)on Si substrate by interface engineering

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Advanced Ceramics 2024年第3期13卷 282-292页

作者： Se hyun Kim Younghwan lee Dong hyun lee Geun Hyeong Park hyun Woo Jeong Kun Yang Yong Hyeon Cho Young Yong Kim Min Hyuk Park Department of Materials Science and Engineering College of EngineeringSeoul National UniversitySeoul 08826Republic of Korea Research Institute of Advanced Materials College of EngineeringSeoul National UniversitySeoul 08826Republic of Korea Inter-university Semiconductor Research Center College of EngineeringSeoul National UniversitySeoul 08826Republic of Korea Pohang Accelerator Laboratory Pohang University of Science&TechnologyPohang 790784Republic of Korea

(Hf,Zr)O_(2)offers considerable potential for next-generation semiconductor devices owing to its nonvolatile spontaneous polarization at the nanoscale.However,scaling this material to sub-5 nm thickness poses several challenges,including the formation of an interfacial layer and high trap concentration.In particular,a low-k SiO_(2)interfacial layer is naturally formed when(Hf,Zr)O_(2)films are directly grown on a Si substrate,leading to high depolarization fields and rapid reduction of the remanent polarization.To address these issues,we conducted a study to significantly improve ferroelectricity and switching endurance of(Hf,Zr)O_(2)films with sub-5 nm thicknesses by inserting a TiO_(2)interfacial layer.The deposition of a Ti film prior to Hf_(0.5)Zr_(0.5)O_(2)film deposition resulted in a high-k TiO_(2)interfacial layer and prevented the direct contact of Hf_(0.5)Zr_(0.5)O_(2)with Si.Our findings show that the high-k TiO_(2)interfacial layer can reduce the SiO_(2)/Si interface trap density and the depolarization field,resulting in a switchable polarization of 60.2μC/cm^(2)for a 5 nm thick Hf_(0.5)Zr_(0.5)O_(2)film.Therefore,we propose that inserting a high-k TiO_(2)interfacial layer between the Hf_(0.5)Zr_(0.5)O_(2)film and the Si substrate may offer a promising solution to enhancing the ferroelectricity and reliability of(Hf,Zr)O_(2)grown on the Si substrate and can pave the way for next-generation semiconductor devices with improved performance.

关键词： ferroelectric charge trapping hafnia depolarization

Chemical and mechanical properties of stainless, environment-friendly, and nonflammable Mg alloys (SEN alloys): A review

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Journal of Magnesium and Alloys 2024年第3期12卷 841-872页

作者： Jong Un lee hyun Ji Kim sang-Cheol Jin Ye Jin Kim Young Min Kim Bong Sun You Jun Ho Bae Sung Hyuk Park Advanced Metals Division Korea Institute of Materials ScienceChangwon 51508Republic of Korea Department of Materials Science and Metallurgical Engineering Kyungpook National UniversityDaegu 41566Republic of Korea Missile Research Institute Agency for Defense DevelopmentDaejeon 34060Republic of Korea

This review article provides overall understanding of stainless,environment-friendly,and nonflammable Mg alloys(SEN alloys)recently developed at the Korea Institute of Materials Science.SEN alloys are produced by adding small amounts of Ca and Y(each<1 wt%)into commercial Mg–Al based alloys,resulting in exceptional ignition and corrosion resistances and impressive mechanical properties.Their main advantages of SEN alloys are as follows.(1)A dense multi-oxide layer of SEN alloys comprising MgO,CaO,and Y_(2)O_(3) impedes the outward dispersion of Mg vapor and the inward penetration of O_(2) during oxidation,thereby enhancing the oxidation and ignition resistances.(2)The presence of Ca-and Y-based second-phase particles in SEN alloys can enhance their corrosion resistance because Ca-containing particles prevent the spread of corrosion,and the replacement of Al-containing particles with less noble ones containing Y(e.g.,Al–Mn–Y or Al–Y particles)retards corrosion.(3)The addition of minor amounts of Ca and Y renders excellent mechanical properties due to improved strengthening effects.These enhanced properties are attributed to more pronounced dynamic recrystallization and grain refining behaviors caused by the second-phase particles during extrusion.(4)Despite the presence of various types of second-phase particles,the fatigue properties of SEN9 alloys are similar to those of commercial AZ91 alloys.(5)Simultaneous introduction of Ca and Y suppresses the formation of Mg17Al12 discontinuous precipitates during aging,leading to the enhanced elongation of aged SEN alloys.(6)Adding mischmetal into the SEN9 alloy leads to a six-fold enhancement in extrudability.Consequently,the studies conducted on SEN alloys demonstrate their excellent ignition and corrosion resistances and mechanical properties,which broaden the industrial applications of Mg alloys by addressing their inherent weaknesses.

关键词： SEN magnesium alloy Corrosion resistance Ignition resistance Mechanical properties Extrudability.