检索结果-南通市图书馆

Machine Intelligence Research 2024年第1期21卷 89-103页

作者： Chaojie wu Mingyang Li Ying Gao Xinyan Xie wing w.Y.ng Ahmad Musyafa School of Computer Science and Engineering South China University of TechnologyGuangzhou 511442China Guangdong provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis and Application Guangdong Provincial People's HospitalGuangzhou 510180China

The problem of art forgery and infringement is becoming increasingly prominent,since diverse self-media contents with all kinds of art pieces are released on the Internet every day.For art paintings,object detection and localization provide an efficient and ef-fective means of art authentication and copyright protection.However,the acquisition of a precise detector requires large amounts of ex-pensive pixel-level annotations.To alleviate this,we propose a novel weakly supervised object localization(wSOL)with background su-perposition erasing(BSE),which recognizes objects with inexpensive image-level labels.First,integrated adversarial erasing(IAE)for vanilla convolutional neural network(CNN)dropouts the most discriminative region by leveraging high-level semantic information.Second,a background suppression module(BSM)limits the activation area of the IAE to the object region through a self-guidance mechanism.Finally,in the inference phase,we utilize the refined importance map(RIM)of middle features to obtain class-agnostic loc-alization results.Extensive experiments are conducted on paintings,CUB-200-2011 and ILSVRC to validate the effectiveness of our BSE.

关键词： weakly supervised object localization erasing method deep learning computer vision art authentication and copyright protection

Low-rank matrix recovery with total generalized variation for defending adversarial examples

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Frontiers of Information Technology & Electronic Engineering 2024年第3期25卷 432-445页

作者： wen LI Hengyou wAng Lianzhi HUO Qiang HE Linlin CHEN Zhiquan HE wing w.Y.ng School of Science Beijing University of Civil Engineering and ArchitectureBeijing 100044China School of Computer Science and Engineering South China University of TechnologyGuangzhou 510006China Aerospace Information Research Institute Chinese Academy of SciencesBeijing 100094China Guangdong Key Laboratory of Intelligent Information processing Shenzhen 518060China Institute of Big Data Modeling and Technology Beijing University of Civil Engineering and ArchitectureBeijing 100044China

Low-rank matrix decomposition with first-order total variation(TV)regularization exhibits excellent performance in exploration of image structure.Taking advantage of its excellent performance in image denoising,we apply it to improve the robustness of deep neural networks.However,although TV regularization can improve the robustness of the model,it reduces the accuracy of normal samples due to its over-smoothing.In our work,we develop a new low-rank matrix recovery model,called LRTGV,which incorporates total generalized variation(TGV)regularization into the reweighted low-rank matrix recovery model.In the proposed model,TGV is used to better reconstruct texture information without over-smoothing.The reweighted nuclear norm and Li-norm can enhance the global structure information.Thus,the proposed LRTGV can destroy the structure of adversarial noise while re-enhancing the global structure and local texture of the image.To solve the challenging optimal model issue,we propose an algorithm based on the alternating direction method of multipliers.Experimental results show that the proposed algorithm has a certain defense capability against black-box attacks,and outperforms state-of-the-art low-rank matrix recovery methods in image restoration.

关键词： Total generalized variation Low-rank matrix Alternating direction method of multipliers Adversarial example

Dysfunction of synaptic endocytic trafficking in parkinson's disease

在线全文

学校读者我要写书评

暂无评论

Neural Regeneration Research 2024年第12期19卷 2649-2660页

作者： Xin Yi ng Mian Cao programme in Neuroscience and Behavioural Disorders Duke-NUS Medical SchoolSingaporeSingapore Department of physiology National University of SingaporeSingaporeSingapore

parkinson's disease is characterized by the selective degeneration of dopamine neurons in the nigrostriatal pathway and dopamine deficiency in the striatum.The precise reasons behind the specific degeneration of these dopamine neurons remain largely elusive.Genetic investigations have identified over 20 causative pARK genes and 90 genomic risk loci associated with both familial and sporadic parkinson's disease.Notably,several of these genes are linked to the synaptic vesicle recycling process,particularly the clathrinmediated endocytosis pathway.This suggests that impaired synaptic vesicle recycling might represent an early feature of parkinson's disease,followed by axonal degeneration and the eventual loss of dopamine cell bodies in the midbrain via a"dying back"mechanism.Recently,several new animal and cellular models with parkinson's disease-linked mutations affecting the endocytic pathway have been created and extensively characterized.These models faithfully recapitulate certain parkinson's disease-like features at the animal,circuit,and cellular levels,and exhibit defects in synaptic membrane trafficking,further supporting the findings from human genetics and clinical studies.In this review,we will first summarize the cellular and molecular findings from the models of two parkinson's disease-linked clathrin uncoating proteins:auxilin(DNAJC6/pARK19)and synaptojanin 1(SYNJ1/pARK20).The mouse models carrying these two pARK gene mutations phenocopy each other with specific dopamine terminal pathology and display a potent synergistic effect.Subsequently,we will delve into the involvement of several clathrin-mediated endocytosis-related proteins(GAK,endophilin A1,SAC2/INpp5 F,synaptotagmin-11),identified as parkinson's disease risk factors through genome-wide association studies,in parkinson's disease pathogenesis.we will also explore the direct or indirect roles of some common parkinson's disease-linked proteins(alpha-synuclein(pARK1/4),parkin(pARK2),and LRRK2(pARK8))in synaptic endocytic trafficking.Additionall

关键词： autophagy auxilin/pARK19 clathrin-mediated endocytosis dopamine neurons neurodegeneration nigrostriatal pathway parkinson's disease synaptic vesicle recycling synaptojanin1/pARK20

Developing a two-decadal time-record of rice field maps using Landsat-derived multi-index image collections with a random forest classifier:A Google Earth Engine based approach

在线全文

学校读者我要写书评

暂无评论

Information processing in Agriculture 2024年第2期11卷 260-275页

作者： w.Ashane M.Fernando I.p.Senanayake Department of Nano Science Technology Wayamba University of Sri LankaKuliyapitiya 60200Sri Lanka Dchoolf Engineing CllegeoEngineeringSience and EnionmentThe UniversyewcastleCallaghanW2Austraia

Historic maps showing the temporal distribution of rice fields are important for precision agriculture,irrigation optimisation,forecasting crop yields,land use management and formulating policies.However,mapping rice felds using traditional ground surveys is impractical when high cost,time and labour requirements are considered,and the availability of such detailed records is limited.Although satellite remote sensing appears to be a viable solution,conventional segmentation and classification methods with spectral bands are often unable to contrast the distinct characteristics between rice fields and other vegetation classes.To this end,we explored a novel,Google Earth Engine(GEE)based multiindex random forest(RF)classification approach to map rice fields over two decades.Landsat images from 2000 to 2020 of two Sri Lankan rice cultivation districts were extracted from GEE and a multi-index RF classification algorithm was applied to distinguish the rice fields.The results showed above 80%accuracy for both training and validation,when compared against high spatial resolution Google Earth imagery.In essence,multi-index sampling and RF together synergised the compelling classifcation accuracy by effectively capturing vegetation,water(ponding)and soil characteristics unique to the rice felds using a single-click approach.The maps developed in this study were further compared against the MODIS land cover type product(MCD12Q1)and the corresponding superior statistics on rice fields demonstrated the robustness of the proposed approach.Future work seeking effective index combinations is recommended,and this approach can potentially be extended to other crop analyses elsewhere.

关键词： Google Earth Engine(GEE) Image classification Random forest Mapping rice fields Time series analysis Vegetation index

Superplasticity of fine-grained Mg-10Li alloy prepared by severe plastic deformation and understanding its deformation mechanisms

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1期12卷 316-331页

作者： H.T.Jeong S.w.Lee w.J.Kim Department of Materials Science and Engineering Hongik UniversityMapo-guSangsu-dong 72-1Seoul 121-791Republic of Korea

The superplastic behavior and associated deformation mechanisms of a fine-grained Mg-10.1 Li-0.8Al-0.6Zn alloy(LAZ1011)with a grain size of 3.2μm,primarily composed of the BCCβphase and a small amount of the HCpαphase,were examined in a temperature range of 473 K to 623 K.The microstructural refinement of this alloy was achieved by employing high-ratio differential speed rolling.The best superplasticity was achieved at 523 K and at strain rates of 10^(-4)-5×10^(-4)s^(-1),where tensile elongations of 550±600%were obtained.During the heating and holding stage of the tensile samples prior to tensile loading,a significant increase in grain size was observed at temperatures above 573 K.Therefore,it was important to consider this effect when analyzing and understanding the superplastic deformation behavior and mechanisms.In the investigated strain rate range,the superplastic flow at low strain rates was governed by lattice diffusion-controlled grain boundary sliding,while at high strain rates,lattice diffusion-controlled dislocation climb creep was the rate-controlling deformation mechanism.It was concluded that solute drag creep is unlikely to occur.During the late stages of deformation at 523 K,it was observed that grain boundary sliding led to the agglomeration of theαphase,resulting in significant strain hardening.Deformation mechanism maps were constructed forβ-Mg-Li alloys in the form of 2D and 3D formats as a function of strain rate,stress,temperature,and grain size,using the constitutive equations for various deformation mechanisms derived based on the data of the current tests.

关键词： Magnesium-lithium alloy Superplasticity Severe plastic deformation Grain size Grain growth

Integrins and their potential roles in mammalian pregnancy

在线全文

学校读者我要写书评

暂无评论

Journal of Animal Science and Biotechnology 2024年第1期15卷 1-19页

作者： Gregory A.Johnson Robert C.Burghardt Fuller w.Bazer Heewon Seo Joe w.Cain Department of Veterinary Integrative Biosciences School of Veterinary Medicine and Biomedical SciencesTexas A&M UniversityCollege StationTX 77843‑4459USA Department of Animal Science College of Agriculture and Life SciencesTexas A&M UniversityCollege StationTX 77843‑2471USA

Integrins are a highly complex family of receptors that, when expressed on the surface of cells, can mediate reciprocal cell-to-cell and cell-to-extracellular matrix(ECM) interactions leading to assembly of integrin adhesion complexes(IACs) that initiate many signaling functions both at the membrane and deeper within the cytoplasm to coordinate processes including cell adhesion, migration, proliferation, survival, differentiation, and metabolism. All metazoan organisms possess integrins, and it is generally agreed that integrins were associated with the evolution of multicellularity, being essential for the association of cells with their neighbors and surroundings, during embryonic development and many aspects of cellular and molecular biology. Integrins have important roles in many aspects of embryonic development, normal physiology, and disease processes with a multitude of functions discovered and elucidated for integrins that directly influence many areas of biology and medicine, including mammalian pregnancy, in particular implantation of the blastocyst to the uterine wall, subsequent placentation and conceptus(embryo/fetus and associated placental membranes) development. This review provides a succinct overview of integrin structure, ligand binding, and signaling followed with a concise overview of embryonic development, implantation, and early placentation in pigs, sheep, humans, and mice as an example for rodents. A brief timeline of the initial localization of integrin subunits to the uterine luminal epithelium(LE) and conceptus trophoblast is then presented, followed by sequential summaries of integrin expression and function during gestation in pigs, sheep, humans, and rodents. As appropriate for this journal, summaries of integrin expression and function during gestation in pigs and sheep are in depth, whereas summaries for humans and rodents are brief. Because similar models to those illustrated in Fig. 1, 2, 3, 4, 5 and 6 are present throughout the scientific literature, the illustrations in this m

关键词： Humans Implantation Integrins pigs pregnancy Rodents Sheep

pervasive Attentive Neural Network for Intelligent Image Classification Based on N-CDE’s

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 1137-1156页

作者： Anas w.Abulfaraj Department of Information Systems Faculty of Computing and Information TechnologyKing Abdulaziz UniversityRabigh21911Saudi Arabia

The utilization of visual attention enhances the performance of image classification tasks.previous attentionbased models have demonstrated notable performance,but many of these models exhibit reduced accuracy when confronted with inter-class and intra-class similarities and differences.Neural-Controlled Differential Equations(N-CDE’s)and Neural Ordinary Differential Equations(NODE’s)are extensively utilized within this context.NCDE’s possesses the capacity to effectively illustrate both inter-class and intra-class similarities and differences with enhanced clarity.To this end,an attentive neural network has been proposed to generate attention maps,which uses two different types of N-CDE’s,one for adopting hidden layers and the other to generate attention values.Two distinct attention techniques are implemented including time-wise attention,also referred to as bottom N-CDE’s;and element-wise attention,called topN-CDE’s.Additionally,a trainingmethodology is proposed to guarantee that the training problem is sufficiently presented.Two classification tasks including fine-grained visual classification andmulti-label classification,are utilized to evaluate the proposedmodel.The proposedmethodology is employed on five publicly available datasets,including CUB-200-2011,ImageNet-1K,pASCAL VOC 2007,pASCAL VOC 2012,and MS COCO.The obtained visualizations have demonstrated that N-CDE’s are better appropriate for attention-based activities in comparison to conventional NODE’s.

关键词： Differential equations neural-controlled DE image classification attention maps N-CDE’s

Cultivating potential:Harnessing plant stem cells for agricultural crop improvement

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Molecular plant 2024年第1期17卷 50-74页

作者： penelope Lindsay Kyle w.Swentowsky David Jackson Cold Spring Harbor Laboratory Cold Spring HarborNY 11724USA National Key Laboratory of Crop Genetic Improvement Huazhong Agricultural UniversityWuhanHubei430070China

Meristems are stem cell-containing structures that produce all plant organs and are therefore important targets for crop improvement.Developmental regulators control the balance and rate of cell divisions within the meristem.Altering these regulators impacts meristem architecture and,as a consequence,plant form.In this review,we discuss genes involved in regulating the shoot apical meristem,inflorescence meristem,axillary meristem,root apical meristem,and vascular cambium in plants.we highlight several examples showing how crop breeders have manipulated developmental regulators to modify meristem growth and alter crop traits such as inflorescence size and branching patterns.plant transformation techniques are another innovation related to plant meristem research because they make crop genome engineering possible.we discuss recent advances on plant transformation made possible by studying genes controlling meristem development.Finally,we conclude with discussions about how meristem research can contribute to crop improvement in the coming decades.

关键词： meristem crop improvement plant transformation plant breeding crop yield gene editing

The quasi-Gramian solution of a non-commutative extension of the higher-order nonlinear Schr?dinger equation

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Communications in Theoretical physics 2024年第3期76卷 51-61页

作者： H w A Riaz J Lin Department of physics Zhejiang Normal UniversityJinhua 321004China

The nonlinear Schr?dinger(NLS)equation,which incorporates higher-order dispersive terms,is widely employed in the theoretical analysis of various physical phenomena.In this study,we explore the non-commutative extension of the higher-order NLS equation.we treat real or complex-valued functions,such as g_(1)=g_(1)(x,t)and g_(2)=g_(2)(x,t)as non-commutative,and employ the Lax pair associated with the evolution equation,as in the commutation case.we derive the quasi-Gramian solution of the system by employing a binary Darboux transformation.The soliton solutions are presented explicitly within the framework of quasideterminants.To visually understand the dynamics and solutions in the given example,we also provide simulations illustrating the associated profiles.Moreover,the solution can be used to study the stability of plane waves and to understand the generation of periodic patterns within the context of modulational instability.

关键词： integrable systems Darboux transformation solitons

Unveiling a new regulator: Vacuolar V-ATpase mediates brassinosteroid signaling in Arabidopsis

在线全文

学校读者我要写书评

暂无评论

Molecular plant 2024年第2期17卷 227-229页

作者： Shikha Malik Justin w.walley Department of plant pathology Entomologyand Microbiologylowa State UniversityAmesIA 50011USA

plant growth and development are intricately regulated by hormones such as auxins,gibberellins,jamonates,cytokinins,and brassinosteroids(BRs).BRs are a group of plant steroid hormones that play a vital role in plant growth and response to the environment.BRs regulate many biological processes including cell elongation,cell division,and male fertility,as well as abiotic and biotic stress responses.

关键词： media fertility brassinosteroid