检索结果-南通市图书馆

Meta label associated loss for fine-grained visual recognition

维普期刊数据库博看期刊评论

在线全文

学校读者我要写书评

暂无评论

Science China(Information Sciences) 2024年第6期67卷 230-247页

作者： Yanchao li fu XIAO Hao li Qun li Shui YU School of Computer Science Nanjing University of Posts and Telecommunications School of Network Engineering Zhoukou Normal University School of Computer Science University of Technology Sydney

Recently, intensive attempts have been made to design robust models for fine-grained visual recognition, most notably are the impressive gains for training with noisy labels by incorporating a reweighting strategy into a meta-learning framework. However, it is limited to up or downweighting the contribution of an instance for label reweighting approaches in the learning process. To solve this issue, a novel noise-tolerant method with auxiliary web data is proposed. Specifically, first, the associations made from embeddings of well-labeled data with those of web data and back at the same class are measured. Next, its association probability is employed as a weighting fusion strategy into angular margin-based loss, which makes the trained model robust to noisy datasets. To reduce the influence of the gap between the well-labeled and noisy web data, a bridge schema is proposed via the corresponding loss that encourages the learned embeddings to be coherent. Lastly, the formulation is encapsulated into the meta-learning framework, which can reduce the overfitting of models and learn the network parameters to be noise-tolerant. Extensive experiments are performed on benchmark datasets, and the results clearly show the superiority of the proposed method over existing state-of-the-art approaches.

关键词： label associated loss weighting noisy samples fine-grained visual recognition noise-tolerant learning meta-learning

Trace formula of the integro-differential operator on a quantum graph

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Applied Mathematics(A Journal of Chinese Universities) 2024年第2期39卷 311-318页

作者： WEI li-xiao YANG Chuan-fu School of Mathematics and Statistics Nanjing University of Science and TechnologyNanjing 210094China.

In this paper we study the eigenvalue problem for integro-differential operators on a lasso graph.The trace formula of the operator is established by applying the residual technique in complex analysis.

关键词： integro-differential operators eigenvalue trace formula

Analysis of amino acid enantiomers in ambient aerosols:Effects and removal of coexistent aerosol matrix

在线全文

学校读者我要写书评

暂无评论

Journal of Environmental Sciences 2024年第3期137卷 732-740页

作者： Ying li Xiaoying li libin Wu Luhan Shi ShanWang Pingqing fu Yingyi Zhang Senchao Lai The Key Lab of Pollution Control and Ecosystem Restoration in Industry Clusters Ministry of EducationSchool of Environment and EnergySouth China University of TechnologyGuangzhou 510006China Institute of Surface-Earth System Science School of Earth System ScienceTianjin UniversityTianjin 300072China now at Hong Kong University of Science and Technology Hong Kong 00852China

Amino acids(AAs)including D-and L-enantiomers are a group of organic nitrogen species in ambient aerosol.Due to the low abundances of AAs(level of ng/m^(3))and the matrix effects by coexistent components,it is challenging to quantify AA enantiomers in ambient aerosols especially under pollution conditions.In this study,we present an optimized method for analyzing AA enantiomers in atmospheric aerosol samples including a pretreatment process and the detection by high performance liquid chromatography coupled to a fuorescence detector(HPLC-FLD).Matrix effects caused by coexistent chemicals on AA enantiomers analysis in ambient aerosol samples were investigated especially for those collected in pollution episodes.The results revealed that the determination of AA enantiomers is significantly affected by the coexistent organic carbon(as a proxy of organic matter)and water-soluble ion of NH_(4)^(+).To remove the matrix effects,we applied a pretreatment using the solid phase extraction column coupled with alkaline adjustment to sample extract.After pretreatment,18 AAs including 6 pairs of D-and L-enantiomers(i.e.,leucine,isoleucine,valine,alanine,serine,and aspartic acid)can be successfully separated and quantified in aerosol samples by HPLC-FLD.The recoveries are in the range of 67%-106%.This method was successfully applied to the urban aerosol samples from pollution and non-pollution periods for AA enantiomers determination.We suggest that the concentrations of D-AAs and the ratio of D-AA/L-AA are indicative of the contribution of bacterial sources and the infuence of biomass burning.

关键词： Amino acid enantiomers Atmospheric aerosol Organic carbon Ammonia Chiral ratio

Tissue adhesive,ROS scavenging and injectable PRP-based‘plasticine’for promoting cartilage repair

在线全文

学校读者我要写书评

暂无评论

Regenerative Biomaterials 2024年第1期3卷 94-107页

作者： Shiao li Dawei Niu Haowei Fang Yancheng Chen Jinyan li Kunxi Zhang Jingbo Yin Peiliang fu Department of Orthopedics Shanghai Changzheng HospitalNaval Medical UniversityShanghai 200003P.R.China Department of Polymer Materials School of Materials Science and EngineeringShanghai UniversityShanghai 200444P.R.China

Platelet-rich plasma(PRP)that has various growth factors has been used clinically in cartilage repair.However,the short residence time and release time at the injury site limit its therapeutic effect.The present study fabricated a granular hydrogel that was assembled from gelatin microspheres and tannic acid through their abundant hydrogen bonding.Gelatin microspheres with the gelatin concentration of 10wt%and the diameter distribution of 1-10 lm were used to assemble by tannic acid to form the granular hydrogel,which exhibited elasticity under low shear strain,but flowability under higher shear strain.The viscosity decreased with the increase in shear rate.Meanwhile,the granular hydrogel exhibited self-healing feature during rheology test.Thus,granular hydrogel carrying PRP not only exhibited well-performed injectability but also performed like a‘plasticine’that possessed good plasticity.The granular hydrogel showed tissue adhesion ability and reactive oxygen species scavenging ability.Granular hydrogel carrying PRP transplanted to full-thickness articular cartilage defects could integrate well with native cartilage,resulting in newly formed cartilage articular fully filled in defects and well-integrated with the native cartilage and subchondral bone.The unique features of the present granular hydrogel,including injectability,plasticity,porous structure,tissue adhesion and reactive oxygen species scavenging provided an ideal PRP carrier toward cartilage tissue engineering.

关键词： cartilage regeneration granular hydrogel gelatin tannic acid PRP

Reaction pathways and selectivity in the chemo-catalytic conversion of cellulose and its derivatives to ethylene glycol:A review

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Chinese Journal of Chemical Engineering 2024年第2期66卷 310-331页

作者： Yao li Yuchun Zhang Zhiyu li Huiyan Zhang Peng fu College of Agricultural Engineering and Food Science Shandong University of TechnologyZibo 255000China Key Laboratory of Energy Thermal Conversion and Control Ministry of EducationSchool of Energy and EnvironmentSoutheast UniversityNanjing 210096China

Biomass-to-ethylene glycol is an effective means to achieve high-value utilisation of cellulose but is hindered by low conversion efficiency and poor catalyst activity and stability.Glucose and cellobiose are derivatives of cellulose conversion to ethylene glycol,and it is found that studying the reaction process of both can help to understand the reaction mechanism of cellulose.It is desirable to develop a reusable,highly active catalyst to convert cellulose into ethylene glycol.This ideal catalyst might have one or more active sites described the conversion steps above.Here,we discuss the catalyst development of celluloseto-ethylene glycol,including tungsten,tin,lanthanide,and other transition metal catalysts,and special attention is given to the reaction mechanism and kinetics for preparing ethylene glycol from cellulose,and the economic advantages of biomass-to-ethylene glycol are briefly introduced.The insights given in this review will facilitate further development of efficient catalysts,for addressing the global energy crisis and climate change related to the use of fossil fuels.

关键词： Ethylene glycol Cellulose Catalyst Retro-aldol condensation Hydrolysis Kinetics

维普期刊数据库博看期刊

Naturally Crosslinked Biocompatible Carbonaceous liquid Metal Aqueous Ink Printing Wearable Electronics for Multi-Sensing and Energy Harvesting

在线全文

学校读者我要写书评

暂无评论

Nano-Micro Letters 2024年第8期16卷 148-168页

作者： King Yan Chung Bingang Xu Di Tan Qingjun Yang Zihua li Hong fu Nanotechnology Center School of Fashion and TextilesThe Hong Kong Polytechnic UniversityHong Kong 999077People’s Republic of China Department of Mathematics and Information Technology The Education University of Hong KongHong KongPeople’s Republic of China

Achieving flexible electronics with comfort and durability comparable to traditional textiles is one of the ultimate pursuits of smart wearables.Ink printing is desirable for e-textile development using a simple and inexpensive process.However,fabricating high-performance atop textiles with good dispersity,stability,biocompatibility,and wearability for high-resolution,large-scale manufacturing,and practical applications has remained challenging.Here,waterbased multi-walled carbon nanotubes(MWCNTs)-decorated liquid metal(LM)inks are proposed with carbonaceous gallium–indium micro-nanostructure.With the assistance of biopolymers,the sodium alginate-encapsulated LM droplets contain high carboxyl groups which non-covalently crosslink with silk sericin-mediated MWCNTs.E-textile can be prepared subsequently via printing technique and natural waterproof triboelectric coating,enabling good flexibility,hydrophilicity,breathability,wearability,biocompatibility,conductivity,stability,and excellent versatility,without any artificial chemicals.The obtained e-textile can be used in various applications with designable patterns and circuits.Multi-sensing applications of recognizing complex human motions,breathing,phonation,and pressure distribution are demonstrated with repeatable and reliable signals.Self-powered and energy-harvesting capabilities are also presented by driving electronic devices and lighting LEDs.As proof of concept,this work provides new opportunities in a scalable and sustainable way to develop novel wearable electronics and smart clothing for future commercial applications.

关键词： Biocompatible Conductive ink Biopolymer E-textile Carbonaceous liquid metal

MSC-YOLO:Improved YOLOv7 Based on Multi-Scale Spatial Context for Small Object Detection in UAV-View

在线全文

学校读者我要写书评

暂无评论

Computers, Materials & Continua 2024年第4期79卷 983-1003页

作者： Xiangyan Tang Chengchun Ruan Xiulai li Binbin li Cebin fu School of Computer Science and Technology Hainan UniversityHaikou570228China Hainan Blockchain Technology Engineering Research Center Hainan UniversityHaikou570228China School of Cyberspace Security(School of Cryptology) Hainan UniversityHaikou570228China

Accurately identifying small objects in high-resolution aerial images presents a complex and crucial task in thefield of small object detection on unmanned aerial vehicles(UAVs).This task is challenging due to variations inUAV flight altitude,differences in object scales,as well as factors like flight speed and motion blur.To enhancethe detection efficacy of small targets in drone aerial imagery,we propose an enhanced You Only Look Onceversion 7(YOLOv7)algorithm based on multi-scale spatial context.We build the MSC-YOLO model,whichincorporates an additional prediction head,denoted as P2,to improve adaptability for small objects.We replaceconventional downsampling with a Spatial-to-Depth Convolutional Combination(CSPDC)module to mitigatethe loss of intricate feature details related to small objects.furthermore,we propose a Spatial Context Pyramidwith Multi-Scale Attention(SCPMA)module,which captures spatial and channel-dependent features of smalltargets acrossmultiple scales.This module enhances the perception of spatial contextual features and the utilizationof multiscale feature information.On the Visdrone2023 and UAVDT datasets,MSC-YOLO achieves remarkableresults,outperforming the baseline method YOLOv7 by 3.0%in terms ofmean average precision(mAP).The MSCYOLOalgorithm proposed in this paper has demonstrated satisfactory performance in detecting small targets inUAV aerial photography,providing strong support for practical applications.

关键词： Small object detection YOLOv7 multi-scale attention spatial context

A semantic segmentation-based underwater acoustic image transmission framework for cooperative SLAM

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Defence Technology（防务技术） 2024年第3期33卷 339-351页

作者： Jiaxu li Guangyao Han Shuai Chang Xiaomei fu School of Marine Science and Technology Tianjin UniversityTianjin 300072China

With the development of underwater sonar detection technology,simultaneous localization and mapping(SLAM)approach has attracted much attention in underwater navigation field in recent years.But the weak detection ability of a single vehicle limits the SLAM performance in wide areas.Thereby,cooperative SLAM using multiple vehicles has become an important research direction.The key factor of cooperative SLAM is timely and efficient sonar image transmission among underwater vehicles.However,the limited bandwidth of underwater acoustic channels contradicts a large amount of sonar image data.It is essential to compress the images before transmission.Recently,deep neural networks have great value in image compression by virtue of the powerful learning ability of neural networks,but the existing sonar image compression methods based on neural network usually focus on the pixel-level information without the semantic-level information.In this paper,we propose a novel underwater acoustic transmission scheme called UAT-SSIC that includes semantic segmentation-based sonar image compression(SSIC)framework and the joint source-channel codec,to improve the accuracy of the semantic information of the reconstructed sonar image at the receiver.The SSIC framework consists of Auto-Encoder structure-based sonar image compression network,which is measured by a semantic segmentation network's residual.Considering that sonar images have the characteristics of blurred target edges,the semantic segmentation network used a special dilated convolution neural network(DiCNN)to enhance segmentation accuracy by expanding the range of receptive fields.The joint source-channel codec with unequal error protection is proposed that adjusts the power level of the transmitted data,which deal with sonar image transmission error caused by the serious underwater acoustic channel.Experiment results demonstrate that our method preserves more semantic information,with advantages over existing methods at the same compression ratio.It also improves the error

关键词： Semantic segmentation Sonar image transmission Learning-based compression

General Method of Combining Grover and Simon for Attacking Block Ciphers

在线全文

学校读者我要写书评

暂无评论

China Communications 2024年第6期21卷 237-245页

作者： fu Xiangqun Bao Wansu Shi Jianhong li Tan Henan Key Laboratory of Quantum Information and Cryptography PLA SSF IEUZhengzhou 450001China Synergetic Innovation Center of Quantum Information and Quantum Physics University of Science and Technology of ChinaHefei 230026China

The compatibility of different quantum algorithms should be considered when these algorithms are combined.In this paper,the method of combining Grover and Simon is studied for the first time,under some preconditions or assumptions.First,we give two preconditions of applying Grover’s algorithm,which ensure that the success probability of finding the marked element is close to 1.Then,based on these two preconditions,it is found out that the success probability of the quantum algorithm for FXconstruction is far less than 1.furthermore,we give the design method of the Oracle function,and then present the general method of combining Grover and Simon algorithm for attacking block ciphers,with success probability close to 1.

关键词： block ciphers post-quantum crypto quantum algorithms