检索结果-南通市图书馆

Steel Surface Defect Detection Using Learnable Memory Vision Transformer

Computers, Materials & Continua 2025年第1期82卷 499-520页

作者： Syed Tasnimul Karim Ayon Farhan Md.Siraj Jia Uddin department of Computer science and Engineering BRAC UniversityDhaka1212Bangladesh department of ai and Big data Endicott CollegeWoosong UniversityDaejeon34606Republic of Korea

This study investigates the application of Learnable Memory Vision Transformers(LMViT)for detecting metal surface flaws,comparing their performance with traditional CNNs,specifically ResNet18 and ResNet50,as well as other transformer-based models including Token to Token ViT,ViT withoutmemory,and Parallel *** awidely-used steel surface defect dataset,the research applies data augmentation and t-distributed stochastic neighbor embedding(t-SNE)to enhance feature extraction and *** techniques mitigated overfitting,stabilized training,and improved generalization *** LMViT model achieved a test accuracy of 97.22%,significantly outperforming ResNet18(88.89%)and ResNet50(88.90%),aswell as the Token to TokenViT(88.46%),ViT without memory(87.18),and Parallel ViT(91.03%).Furthermore,LMViT exhibited superior training and validation performance,attaining a validation accuracy of 98.2%compared to 91.0%for ResNet 18,96.0%for ResNet50,and 89.12%,87.51%,and 91.21%for Token to Token ViT,ViT without memory,and Parallel ViT,*** findings highlight the LMViT’s ability to capture long-range dependencies in images,an areawhere CNNs struggle due to their reliance on local receptive fields and hierarchical feature *** additional transformer-based models also demonstrate improved performance in capturing complex features over CNNs,with LMViT excelling particularly at detecting subtle and complex defects,which is critical for maintaining product quality and operational efficiency in industrial *** instance,the LMViT model successfully identified fine scratches and minor surface irregularities that CNNs often *** study not only demonstrates LMViT’s potential for real-world defect detection but also underscores the promise of other transformer-based architectures like Token to Token ViT,ViT without memory,and Parallel ViT in industrial scenarios where complex spatial relationships are *** research m

关键词： Learnable Memory Vision Transformer(LMViT) Convolutional Neural Networks(CNN) metal surface defect detection deep learning,computer vision image classification learnable memory gradient clipping label smoothing t-SNE visualization

来源：

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

Power synchronization: taming massive diversified serverless functions under power constraints

引用

science China(Information sciences) 2025年第03期68卷 47-65页

作者： Du LIU Lu ZHANG Yechen XU Xinkai WANG Lingyu SUN Yifei PU Xiaofeng HOU Chao LI Minyi GUO department of Computer science and Engineering Shanghai Jiao Tong University ai Chip Center for Emerging Smart Systems The Hong Kong University of Science and Technology

Carbon neutrality has become an important design objective worldwide. However, the on-going shift to cloud-naive era does not necessarily mean energy efficiency. From the perspective of power management, co-hosted serverless functions are difficult to tame. They are lightweight, short-lived applications sensitive to power capping activities. In addition, they exhibit great individual and temporal variability, presenting idiosyncratic performance/power scaling goals that are often at odds with one another. To date, very few proposals exist in terms of tailored power management for serverless platforms. In this work, we introduce power synchronization, a novel yet generic mechanism for managing serverless functions in a power-efficient way. Our insight with power synchronization is that uniform application power behavior enables consistent and uncompromised function operation on shared host machines. We also propose PowerSync, a synchronization-based power management framework that ensures optimal efficiency based on a clear understanding of functions. Our evaluation shows that PowerSync can improve the energy efficiency of functions by up to 16% without performance loss compared to conventional power management strategies.

关键词： serverless computing variability energy efficiency power management power synchronization

来源：

同方期刊数据库评论

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

Deep Learning and Artificial Intelligence-Driven Advanced Methods for Acute Lymphoblastic Leukemia Identification and Classification: A Systematic Review

引用

Computer Modeling in Engineering & sciences 2025年第2期142卷 1199-1231页

作者： Syed Ijaz Ur Rahman Naveed Abbas Sikandar Ali Muhammad Salman Ahmed Alkhayat Jawad Khan Dildar Hussain Yeong Hyeon Gu department of Computer science Islamia CollegePeshawar25120Pakistan department of Information Technology The University of HaripurHaripur22620Pakistan College of Technical Engineering The Islamic UniversityNajaf100986Iraq School of Computing Gachon UniversitySeongnam13120Republic of Korea department of ai and data science Sejong UniversitySeoul05006Republic of Korea

Automatic detection of Leukemia or blood cancer is one of the most challenging tasks that need to be addressed in the healthcare *** of white blood cells(WBCs)in the blood or bone marrow microscopic slide images play a crucial part in early identification to facilitate medical *** Acute Lymphocytic Leukemia(ALL),the most preferred part of the blood or marrow is to be analyzed by the experts before it spreads in the whole body and the condition becomes *** researchers have done a lot of work in this field,to demonstrate a comprehensive analysis few literature reviews have been published focusing on various artificial intelligence-based techniques like machine and deep learning detection of *** systematic review has been done in this article under the PRISMA guidelines which presents the most recent advancements in this *** image segmentation techniques were broadly studied and categorized from various online databases like Google Scholar,science Direct,and PubMed as image processing-based,traditional machine and deep learning-based,and advanced deep learning-based models were *** Neural Networks(CNN)based on traditional models and then the recent advancements in CNN used for the classification of ALL into its subtypes.A critical analysis of the existing methods is provided to offer clarity on the current state of the ***,the paper concludes with insights and suggestions for future research,aiming to guide new researchers in the development of advanced automated systems for detecting life-threatening diseases.

关键词： Acute lymphoblastic bone marrow segmentation classification machine learning deep learning convolutional neural network

来源：

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

ControlVideo: conditional control for one-shot text-driven video editing and beyond

引用

science China(Information sciences) 2025年第03期68卷 150-162页

作者： Min ZHAO Rongzhen WANG Fan BAO Chongxuan LI Jun ZHU department of Computer science and Technology Institute for AI Tsinghua-Bosch Joint ML CenterTsinghua Laboratory of Brain and Intelligence Lab Tsinghua University ShengShu Technology Gaoling School of Artificial Intelligence Renmin University of China Beijing Key Laboratory of Big data Management and Analysis Methods Pazhou Laboratory (Huangpu)

This paper presents ControlVideo for text-driven video editing — generating a video that aligns with a given text while preserving the structure of the source video. Building on a pre-trained text-to-image diffusion model, ControlVideo enhances the fidelity and temporal consistency by incorporating additional conditions(such as edge maps), and fine-tuning the key-frame and temporal attention on the source video-text pair via an in-depth exploration of the design space. Extensive experimental results demonstrate that ControlVideo outperforms various competitive baselines by delivering videos that exhibit high fidelity w.r.t. the source content, and temporal consistency, all while aligning with the text. By incorporating low-rank adaptation layers into the model before training, ControlVideo is further empowered to generate videos that align seamlessly with reference images. More importantly, ControlVideo can be readily extended to the more challenging task of long video editing(e.g., with hundreds of frames), where maintaining long-range temporal consistency is crucial. To achieve this, we propose to construct a fused ControlVideo by applying basic ControlVideo to overlapping short video segments and key frame videos and then merging them by pre-defined weight functions. Empirical results validate its capability to create videos across 140 frames, which is approximately 5.83 to 17.5 times more than what previous studies achieved. The code is available at https://***/thu-ml/controlvideo.

关键词： diffusion models controllable generation text-driven editing video editing long video editing

来源：

同方期刊数据库

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

Factor-adjusted tests for generalized linear models with multimodal data:An application to breast cancer data

引用

science China Mathematics 2025年第2期68卷 447-484页

作者： Dongyu Li Lei Wang School of Statistics and data science KLMDASRLEBPSand LPMCNankai UniversityTianjin 300071China

With the advancement of modern scientific research,multimodal data is increasingly being collected from multiple sources or *** outcomes derived from generalized linear models with high-dimensional and multimodal covariates,we develop two distinct factor-adjusted tests to assess the significance of high-dimensional modality data and specific low-dimensional linear combinations of predictors from one or more modalities,***,we propose a factor-adjusted decorrelated score test to evaluate the significance of a single *** approach simultaneously transforms a high-dimensional test into a fixed low-dimensional one while addressing the impact of high-dimensional nuisance ***,we construct a factor-adjusted Wald test based on partial penalized estimation to assess the significance of certain low-dimensional combinations of variables from one or more *** limiting distributions of these two proposed tests are analyzed under both the null hypothesis and local alternatives to characterize the asymptotic type-I errors and *** finite sample performance of our proposed tests is evaluated through simulations and further demonstrated with a breast cancer dataset.

关键词： decorrelated score factor model high-dimensional data integrative analysis partial penalized Wald test

来源：

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

Asymptotic false discovery control of the Benjamini-Hochberg procedure for pairwise comparisons

引用

science China Mathematics 2025年第1期68卷 223-252页

作者： Weidong Liu Dennis Leung Qi-Man Shao School of Mathematical sciences Shanghai Jiao Tong UniversityShanghai 200240China School of Mathematics and Statistics University of MelbourneVictoria 3010Australia department of Statistics and data science SICMSouthern University of Science and TechnologyShenzhen 518055China

In a one-way analysis-of-variance(ANOVA) model,the number of pairwise comparisons can become large even with a moderate number of *** by this,we consider a regime with a growing number of groups and prove that,when testing pairwise comparisons,the Benjamini-Hochberg(BH) procedure can asymptotically control false discoveries,despite the fact that the involved t-statistics do not exhibit the wellknown positive dependence structure required for exact false discovery rate(FDR) *** Tukey's perspective that the difference between the means of any two groups cannot be exactly zero,our main result provides control over the directional false discovery rate and directional false discovery proportion.A key technical contribution of our work is demonstrating that the dependence among the t-statistics is sufficiently weak to establish the convergence result typically required for asymptotic FDR *** analysis does not rely on conventional assumptions such as normality,variance homogeneity,or a balanced design,thereby offering a theoretical foundation for applications in more general settings.

关键词： BH procedure directional error FDR one-way ANOVA pairwise comparison Studentized twosample t-statistic Cramer-type moderate deviation uniform law of large numbers

来源：

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

Decreased levels of phosphorylated synuclein in plasma are correlated with poststroke cognitive impairment

引用

Neural Regeneration Research 2025年第9期20卷 2598-2610页

作者： Yi Wang Yuning Li Yakun Gu Wei Ma Yuying Guan Mengyuan Guo Qianqian Shao Xunming Ji Jia Liu department of Clinical Laboratory Beijing Bo’ai HospitalChina Rehabilitation Research CenterCapital Medical UniversityBeijingChina Beijing Institute of Brain Disorders Laboratory of Brain DisordersMinistry of Science and TechnologyCollaborative Innovation Center for Brain DisordersBeijing Advanced Innovation Center for Big Data-based Precision MedicineCapital Medical UniversityBeijingChina department of Neurosurgery Xuanwu HospitalCapital Medical UniversityBeijingChina

Poststro ke cognitive impairment is a major secondary effect of ischemic stroke in many patients;however,few options are available for the early diagnosis and treatment of this *** aims of this study were to(1)determine the specific relationship between hypoxic andα-synuclein during the occur of poststroke cognitive impairment and(2)assess whether the serum phosphorylatedα-synuclein level can be used as a biomarker for poststro ke cognitive *** found that the phosphorylatedα-synuclein level was significantly increased and showed pathological aggregation around the cerebral infa rct area in a mouse model of ischemic *** addition,neuronalα-synuclein phosphorylation and aggregation were observed in the brain tissue of mice subjected to chronic hypoxia,suggesting that hypoxia is the underlying cause ofα-synuclein-mediated pathology in the brains of mice with ischemic *** phosphorylatedα-synuclein levels in patients with ischemic stroke were significantly lower than those in healt hy subjects,and were positively correlated with cognition levels in patients with ischemic ***,a decrease in serum high-density lipoprotein levels in stroke patie nts was significantly correlated with a decrease in phosphorylatedα-synuclein *** ischemic stroke mice did not show significant cognitive impairment or disrupted lipid metabolism 14 days after injury,some of them exhibited decreased cognitive function and reduced phosphorylatedα-synuclein *** together,our results suggest that serum phosphorylatedα-synuclein is a potential biomarker for poststroke cognitive impairment.

关键词： biomarker high-density lipoprotein ischemic stroke phosphorylatedα-synuclein poststroke cognitive impairment

来源：

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

Domain generalization with semi-supervised learning for people-centric activity recognition

引用

science China(Information sciences) 2025年第1期68卷 171-188页

作者： Jing LIU Wei ZHU Di LI Xing HU Liang SONG Academy for Engineering & Technology Fudan University Shanghai East-bund Research Institute on Networking Systems of ai School of Optoelectronic Information and Computer Engineering University of Shanghai for Science & Technology

People-centric activity recognition is one of the most critical technologies in a wide range of real-world applications,including intelligent transportation systems, healthcare services, and brain-computer interfaces. Large-scale data collection and annotation make the application of machine learning algorithms prohibitively expensive when adapting to new tasks. One way of circumventing this limitation is to train the model in a semi-supervised learning manner that utilizes a percentage of unlabeled data to reduce the labeling burden in prediction tasks. Despite their appeal, these models often assume that labeled and unlabeled data come from similar distributions, which leads to the domain shift problem caused by the presence of distribution gaps. To address these limitations, we propose herein a novel method for people-centric activity recognition,called domain generalization with semi-supervised learning(DGSSL), that effectively enhances the representation learning and domain alignment capabilities of a model. We first design a new autoregressive discriminator for adversarial training between unlabeled and labeled source domains, extracting domain-specific features to reduce the distribution gaps. Second, we introduce two reconstruction tasks to capture the task-specific features to avoid losing information related to representation learning while maintaining task-specific consistency. Finally, benefiting from the collaborative optimization of these two tasks, the model can accurately predict both the domain and category labels of the source domains for the classification task. We conduct extensive experiments on three real-world sensing datasets. The experimental results show that DGSSL surpasses the three state-of-the-art methods with better performance and generalization.

关键词： activity recognition deep learning domain generalization semi-supervised learning adversarial training

来源：

同方期刊数据库

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

RE-SEGNN:recurrent semantic evidence-aware graph neural network for temporal knowledge graph forecasting

引用

science China(Information sciences) 2025年第2期68卷 103-119页

作者： Wenyu Cai Mengfan LI Xuanhua SHI Yuanxin FAN Quntao ZHU Hai JIN National Engineering Research Center for Big data Technology and System Services Computing Technology and System LabCluster and Grid Computing LabSchool of Computer Science and TechnologyHuazhong University of Science and Technology

Temporal knowledge graph(TKG) reasoning, has seen widespread use for modeling real-world events, particularly in extrapolation settings. Nevertheless, most previous studies are embedded models, which require both entity and relation embedding to make predictions, ignoring the semantic correlations among different entities and relations within the same timestamp. This can lead to random and nonsensical predictions when unseen entities or relations occur. Furthermore, many existing models exhibit limitations in handling highly correlated historical facts with extensive temporal depth. They often either overlook such facts or overly accentuate the relationships between recurring past occurrences and their current counterparts. Due to the dynamic nature of TKG, effectively capturing the evolving semantics between different timestamps can be *** address these shortcomings, we propose the recurrent semantic evidenceaware graph neural network(RE-SEGNN), a novel graph neural network that can learn the semantics of entities and relations simultaneously. For the former challenge, our model can predict a possible answer to missing quadruples based on semantics when facing unseen entities or relations. For the latter problem, based on an obvious established force, both the recency and frequency of semantic history tend to confer a higher reference value for the current. We use the Hawkes process to compute the semantic trend, which allows the semantics of recent facts to gain more attention than those of distant facts. Experimental results show that RE-SEGNN outperforms all SOTA models in entity prediction on 6 widely used datasets, and 5 datasets in relation prediction. Furthermore, the case study shows how our model can deal with unseen entities and relations.

关键词： knowledge graph reasoning temporal knowledge graph Hawkes process semantic evidence

来源：

同方期刊数据库

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

Three-Stage Transfer Learning with AlexNet50 for MRI Image Multi-Class Classification with Optimal Learning Rate

引用

Computer Modeling in Engineering & sciences 2025年第1期142卷 155-183页

作者： Suganya Athisayamani A.Robert Singh Gyanendra Prasad Joshi Woong Cho School of Computing Sastra Deemed to be UniversityThanjavur613401India department of Computational Intelligence SRM Institute of Science and TechnologyKattankulathur603203India department of ai and Software Engineering Kangwon National UniversitySamcheok25913Republic of Korea department of Electronics Information and Communication EngineeringKangwon National UniversitySamcheok25913Republic of Korea

In radiology,magnetic resonance imaging(MRI)is an essential diagnostic tool that provides detailed images of a patient’s anatomical and physiological *** is particularly effective for detecting soft tissue ***,radiologists manually interpret these images,which can be labor-intensive and time-consuming due to the vast amount of *** address this challenge,machine learning,and deep learning approaches can be utilized to improve the accuracy and efficiency of anomaly detection in MRI *** manuscript presents the use of the Deep AlexNet50 model for MRI classification with discriminative learning *** are three stages for learning;in the first stage,the whole dataset is used to learn the *** the second stage,some layers of AlexNet50 are frozen with an augmented dataset,and in the third stage,AlexNet50 with an augmented dataset with the augmented *** method used three publicly available MRI classification datasets:Harvard whole brain atlas(HWBA-dataset),the School of Biomedical Engineering of Southern Medical University(SMU-dataset),and The National Institute of Neuroscience and Hospitals brain MRI dataset(NINS-dataset)for *** hyperparameter optimizers like Adam,stochastic gradient descent(SGD),Root mean square propagation(RMS prop),Adamax,and AdamW have been used to compare the performance of the learning ***-dataset registers maximum classification *** evaluated the performance of the proposed classification model using several quantitative metrics,achieving an average accuracy of 98%.

关键词： MRI tumors classification AlexNet50 transfer learning hyperparameter tuning optimizer

来源：

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

欢迎您,

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：