检索结果-内蒙古大学图书馆

TLSTMF-YOLO: Transfer Learning and Feature Fusion Network for Earthquake-Induced Landslide Detection in Remote Sensing Images

引用

IEEE Transactions on Geoscience and Remote Sensing 2025年 63卷

作者： Meng, Shaoqiang Shi, Zhenming Pirasteh, Saied Liberata Ullo, Silvia Peng, Ming Zhou, Changshi Nunes Goncalves, Wesley Zhang, Limin Tongji University Shanghai Research Institute for Intelligent Autonomous Systems Key Laboratory of Geotechnical and Underground Engineering Ministry of Education Shanghai200092 China Institute of Artificial Intelligence Shaoxing University Zhejiang Shaoxing312000 China Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Geotechnics and Geomatics Tamil Nadu Chennai600 077 India University of Sannio Department of Engineering Benevento82100 Italy Tongji University Shanghai Research Institute for Intelligent Autonomous Systems Shanghai200092 China Federal University of Mato Grosso do Sul Faculty of Computing Campo Grande79070-900 Brazil The Hong Kong University of Science and Technology Department of Civil and Environmental Engineering Hong Kong Hong Kong

Detecting earthquake-induced landslides in remote sensing images is challenging due to the varying sizes of landslides, uneven distribution, and the prevalence of small targets. This study proposes a novel approach, the TLSTMF-YOLO model, which combines a C3-Swin-Transformer and multiscale feature fusion techniques to enhance detection accuracy and efficiency. Key innovations include the use of a convolutional block attention module (CBAM) to improve feature representation, and a bidirectional feature pyramid network (BiFPN) for optimized cross-scale feature fusion. To address data scarcity, a transfer learning strategy is applied, supported by an AdamW optimizer and cosine learning rate strategy for faster convergence. Evaluations on the Jiuzhaigou and Luding landslide datasets demonstrate the model's effectiveness, achieving precision, recall, and mean average precision (mAP)@0.5 of 95.7%, 89.9%, and 90.5% on the Jiuzhaigou dataset, and 96.0%, 90.9%, and 94.5% on the Luding dataset, respectively. In addition, the model processes frames efficiently, with times of 6.61 and 12.2 ms on the two datasets. These results confirm the model's capability for accurate and efficient landslide detection, highlighting its potential for real-world applications. © 1980-2012 IEEE.

关键词： Earthquakes

来源：评论

学校读者我要写书评

暂无评论

Hierarchical progressive learning for zero-shot peptide-HLA binding prediction and automated antigenic peptide design

引用

Cell reports 2025年第6期44卷 115763页

作者： Xinyuan Zhu Jiadong Lu Xinting Hu Tengchuan Jin Shan Lu Fuli Feng School of Information Science and Technology University of Science and Technology of China Hefei Anhui China. School of Artificial Intelligence and Data Science University of Science and Technology of China Hefei Anhui China. School of Computer Science and Engineering Nanyang Technological University Singapore Singapore. Division of Life Sciences and Medicine University of Science and Technology of China Hefei Anhui China. Department of Cellular and Molecular Medicine University of California San Diego San Diego CA USA. Electronic address: ms.shanlu@***. School of Artificial Intelligence and Data Science University of Science and Technology of China Hefei Anhui China. Electronic address: fengfl@***.

Predicting the binding of peptides to human leukocyte antigen (HLA) alleles is critical for identifying epitopes initiating immune responses in vaccine development and immunotherapies. Despite the progress in predicting binding peptides of 1% HLA class-I alleles with in silico prediction tools, it remains challenging to predict binding peptides for alleles lacking epitopes. To achieve the binding prediction on those zero-shot alleles, we developed a hierarchical progressive learning (HPL) framework that progressively learns sequence patterns of specific peptide-HLA complexes. HPL models improve the prediction performance for zero-shot HLA class-I alleles by 60.8% (1,414.0% for non-classical ones). We further developed an automated peptide mutation search program, which automatically identifies binding peptides for any target HLA class-I allele and mutates weak-binding or non-binding peptides to enhance binding affinity under the guidance of HPL models. Our program generated high-affinity binding candidates of any target allele with limited mutations in more than 38.1% of testing cases.

关键词： CP: Immunology HLA-peptide binding immunotherapy protein language model

来源：评论

学校读者我要写书评

暂无评论

Fuzzy Triple Contrastive Learning for Hyperspectral Image Classification

引用

IEEE Transactions on Geoscience and Remote Sensing 2025年 63卷

作者： Yonghe Chu Zhaolong Wang Jiangtao Peng Weiping Ding Heling Cao School of Artificial Intelligence and Computer Science Nantong University Nantong China College of Information Science and Engineering Henan University of Technology Zhengzhou China Hubei Key Laboratory of Applied Mathematics Faculty of Mathematics and Statistics Key Laboratory of Intelligent Sensing System and Security Ministry of Education Hubei University Wuhan China Faculty of Data Science City University of Macau Macau China

Recently, contrastive learning (CL) has shown excellent performance in the hyperspectral image (HSI) classification. However, existing CL-based methods face two specific challenges: 1) multiview samples inevitably introduce ambiguity and uncertainty due to data augmentation operations. Traditional contrastive learning methods fail to effectively model these dynamic ambiguous features, resulting in a lack of robustness in the feature learning process and 2) existing CL-based methods primarily learn feature representations by pulling positive samples closer and pushing negative samples apart. However, they lack structured modeling of intraclass feature compactness and interclass feature separability. To address these challenges, we propose a fuzzy triplet contrastive learning (FTCL) method for HSI classification. For the first challenge, we propose a multiview fuzzy neighborhood learning (MFNL) module. This module effectively models the ambiguity among multiview samples through fuzzy membership calculation, multiview fuzzy weight matrix generation, and weighted feature aggregation, significantly enhancing the robustness and stability of feature representations. To tackle the second challenge, we design a triplet feature discriminative (TFD) classifier, which improves intraclass compactness by minimizing the distance between anchor samples and positive samples, while enhancing interclass separability by maximizing the distance between anchor samples and negative samples. This enables precise modeling of intraclass compactness and interclass separability. The proposed method is evaluated on four HSI datasets, and the experimental results demonstrate that the proposed method outperforms the state-of-the-art methods.

关键词： Contrastive learning Feature extraction Hyperspectral imaging Robustness Training data mining Three-dimensional displays Uncertainty Image classification Deep learning

来源：评论

学校读者我要写书评

暂无评论

Explainable seismic damage prediction model based on CELS-WOA-Stacking

引用

Advanced engineering Informatics 2025年 66卷

作者： Gu, Yi Yang, Shichao Zhou, Xu Liu, Yongkang College of Civil and Architectural Engineering North China University of Science and Technology Tangshan063210 China College of Artificial Intelligence North China University of Science and Technology Tangshan063210 China College of Science North China University of Science and Technology Tangshan063210 China Hebei Key Laboratory of Data Science and Applications Tangshan063210 China College of Electrical Engineering North China University of Science and Technology Tangshan063210 China

Rapid and accurate assessment of post-earthquake building damage is crucial for effective emergency response and recovery. This study aims to enable knowledge-guided decision support in seismic disaster scenarios and to examine the relationships between influencing factors and seismic damage. Based on the Nepal Earthquake Damage database (NBDP), we developed a hybrid earthquake damage prediction model that combines an integrated stacking framework with an improved whale optimization algorithm (Chaotic-Lévy Elite-enhanced Whale Optimization Algorithm, CELS-WOA). To address model complexity in resource-constrained post-disaster environments, we employed committee voting for feature selection and quantified the contribution of each variable using Shapley Additive Explanations (SHAP). The results demonstrate that our model outperforms single models and traditional ensemble approaches under various optimization strategies, achieving a prediction accuracy of 0.904. Additionally, we analyzed the SHAP values of regional variables in conjunction with seismic intensity information, uncovering latent mechanisms of seismic intensity encoding through geospatial attributes. Key indicators affecting seismic damage include pre/post-earthquake building height/floors, geographic location, construction age, and mud-stone material usage. Specifically, building height/floors directly reflect damage, while geographic location influences damage through hidden earthquake intensity. Construction age and mud-stone material influence the damage of the building by reflecting their own ability to resist vibration. This study clarifies the impact of influencing factors on the earthquake loss of buildings based on machine learning, which provides a basis for intelligent loss assessment for earthquake disaster scenarios. © 2025 Elsevier Ltd

关键词： Earthquake effects

来源：评论

学校读者我要写书评

暂无评论

EnsembleNTLDetect: An intelligent framework for electricity theft detection in smart grid

arXiv

引用

arXiv 2021年

作者： Kulkarni, Yogesh Sayf Hussain, Z. Ramamritham, Krithi Somu, Nivethitha Department of Computer Engineering Pune Institute of Computer Technology Pune India Department of Computer Science & Engineering College of Engineering Guindy Anna University Chennai India Robert Bosch Centre for Data Science & Artificial Intelligence Iit Madras Chennai India Department of Electrical Engineering Iit Bombay Mumbai India

artificial intelligence-based techniques applied to the electricity consumption data generated from the smart grid prove to be an effective solution in reducing Non Technical Loses (NTLs), thereby ensures safety, reliability, and security of the smart energy systems. However, imbalanced data, consecutive missing values, large training times, and complex architectures hinder the real time application of electricity theft detection models. In this paper, we present EnsembleNTLDetect, a robust and scalable electricity theft detection framework that employs a set of efficient data pre-processing techniques and machine learning models to accurately detect electricity theft by analysing consumers' electricity consumption patterns. This framework utilises an enhanced Dynamic Time Warping Based Imputation (eDTWBI) algorithm to impute missing values in the time series data and leverages the Near-miss undersampling technique to generate balanced data. Further, stacked autoencoder is introduced for dimensionality reduction and to improve training efficiency. A Conditional Generative Adversarial Network (CTGAN) is used to augment the dataset to ensure robust training and a soft voting ensemble classifier is designed to detect the consumers with aberrant consumption patterns. Furthermore, experiments were conducted on the real-time electricity consumption data provided by the State Grid Corporation of China (SGCC) to validate the reliability and efficiency of EnsembleNTLDetect over the state-of-the-art electricity theft detection models in terms of various quality metrics. © 2021, CC BY-NC-ND.

关键词： Electric power utilization

来源：评论

学校读者我要写书评

暂无评论

EEG signal analysis using deep learning: A systematic literature review 5

EEG signal analysis using deep learning: A systematic litera...

引用

5th International Conference on Intelligent Computing in data sciences, ICDS 2021

作者： Amrani, Ghita Adadi, Amina Berrada, Mohammed Souirti, Zouhayr Boujraf, Said Laboratory of Artificial Intelligence Data Science and Emerging Systems National School of Applied Sciences Fez Morocco Moulay Ismail University ISIC Research Team of High School of Technology LMMI Laboratory Meknes Morocco Sleep Center Hassan II University Hospital Clinical Neurosciences Laboratory Faculty of Medicine and Pharmacy Neurology Department Fez Morocco Hassan II University Hospital Clinical Neurosciences Laboratory Department of Biophysics and Clinical MRI Methods Faculty of Medicine Pharmacy Department of Radiology and Clinical Imaging Fez Morocco

ISBN: (纸本)9781665442381

Objective: Electroencephalography (EEG) is very crucial for understanding the dynamic healthy and pathological complex processes in the brain. However, the manual analysis of the EEG signal is very complex, time-consuming, and depends on the expertise and experience of the users. Hence, nowadays research is conducted on automated EEG signal analysis using artificial intelligence and computer-aided technologies. This would allow fast and highly accurate results. The goal of this paper is to provide an extensive review of the EEG signal analysis using deep learning (DL).Methods: This systematic literature review of EEG processing using Deep Learning (DL) was achieved on Web of science, PubMed, and science Direct databases, resulting in 403 identified papers. All collected studies were analyzed based on main disorders studied, type of tasks performed, data source, stages of analysis, and DL ***: DL in EEG processing is promising in various research applications. It covered the common neurological disorders diagnosis such as epilepsy, movement disorder, depression, schizophrenia, autism, alcohol use, attention, memory, sleep, pain, etc. The main tasks covered by the included studies are detection and classification. The average range of data sources utilized by the included studies is 127 subjects with an EEG recording a total duration of 458 hours. In fact, we identified the use of a plethora of DL architecture for EEG analysis. 57% of papers used Convolutional Neural Networks (CNNs), whereas Recurrent Neural Networks (RNNs) were the architecture choice of about 12% of papers. Combinations of CNNs and Long Short-Term Memory (LSTM) were used in 13% of studies. Generative Adversarial Networks (GAN) and Autoencoder (AEs) were used in 5% and 4% of papers respectively. Restricted Boltzmann Machine (RBMs), Deep Belief Networks (DBNs), and other hybrid architectures appeared in 1% of ***: This systematic review showed that DL is a powerful tool

关键词： Electroencephalography

来源：评论

学校读者我要写书评

暂无评论

Efficient Video Summarization for Smart Surveillance Systems

Efficient Video Summarization for Smart Surveillance Systems

引用

IEEE Symposium Series on Computational intelligence (SSCI)

作者： Khan Muhammad Tanveer Hussain Javier Del Ser Weiping Ding Amir H. Gandomi Victor Hugo C. De Albuquerque Department of Applied Artificial Intelligence Sungkyunkwan University Seoul South Korea Institute for Transport Studies University of Leeds Leeds United Kingdom Department of Communications Engineering University of the Basque Country (UPV/EHU) Bilbao Spain School of Information Science and Technology Nantong University Nantong China Faculty of Engineering & Information Technology University of Technology Sydney Sydney Australia Department of Teleinformatics Engineering Federal University of Ceará Fortaleza Fortaleza CE Brazil

ISBN: (纸本)9781665487696

In surveillance systems, vast amounts of data are collected from different sources to monitor ongoing video activities. Usually, video data is passively captured by visual sensors and forwarded to the command center, without intelligent Edge functionalities to select essential video information and locally detect abnormal events therein. These shortcomings often seen in practical surveillance scenarios lead to a wastage of storage resources and make data management, retrieval, and informed decision complex and time-consuming. Therefore, endowing visual sensors with video summarization capabilities is of utmost importance for smarter surveillance systems. This study departs from this rationale to propose an efficient neural networks-based video summarization method for surveillance systems. The proposed approach learns how to optimally segment a video by measuring informative features from the data flow, followed by memorability and entropy to maintain the relevance and diversity of the video summary produced on the Edge. Experimental results over benchmark datasets reveal that the proposed scheme outperforms other state-of-the-art counterparts and proves the effectiveness of our method for video summarization in smart cities.

关键词： Visualization Smart cities Surveillance Computational modeling Benchmark testing Streaming media Feature extraction

来源：评论

学校读者我要写书评

暂无评论

An enhanced UHMWPE wear particle detection approach based on YOLOv9

引用

Medical engineering & Physics 2025年 142卷

作者： Lingmeng Li Mingzhen Deng Steven Su Richard M. Hall J.L. Tipper Faculty of Engineering and IT University of Technology Sydney NSW Australia College of Artificial Intelligence and Big Data for Medical Science Shandong First Medical University & Shandong Academy of Medical Sciences Taian China Department of Mechanical Engineering University of Birmingham Birmingham United Kingdom STEM College RMIT University Melbourne VIC Australia

Ultra-high molecular weight polyethylene (UHMWPE) has been widely used in total joint arthroplasty for orthopedic and spinal implants. However, the biological response to UHMWPE wear particles has been identified as a major contributor to inflammatory synovitis and periprosthetic osteolysis, which could lead to aseptic loosening and long-term implant failure. Traditional manual detection and classification of UHMWPE wear particles are labor-intensive, time-consuming, and prone to human error, which requires the development of automated detection techniques. This study proposes a novel deep learning-based framework for detecting UHMWPE wear particles, utilizing high-resolution field emission gun-scanning electron microscopy (FEG-SEM) images. The proposed approach employs an enhanced YOLOv9 object detection model, incorporating programmable gradient information (PGI) and generalized efficient layer aggregation networks (GELAN) to improve the localization and detection accuracy of small objects. Additionally, a customized Focal Loss function is integrated to address class imbalance and enhance sensitivity to submicron and nanoscale wear particles. Experimental evaluations demonstrate that our proposed model achieves a mean average precision (mAP) of 84.0%, outperforming the baseline YOLOv5 model by 7.7%. Furthermore, compared to mainstream object detection models such as YOLOv8 and Faster R-CNN, our approach exhibits superior detection accuracy and robustness, particularly in identifying wear particles in complex backgrounds and overlapping regions. In addition to developing an advanced detection algorithm, this study establishes a dedicated and expert-annotated UHMWPE wear particle dataset, addressing a critical gap in orthopedic implant research. The proposed framework provides a scalable, high-precision, and cost-effective solution for the automated detection of UHMWPE wear particles, supporting improved implant monitoring, osteolysis prevention, and clinical decisi

关键词：

来源：评论

学校读者我要写书评

暂无评论

6G communications through sub-Terahertz CMOS power amplifiers: Design challenges and trends

引用

Heliyon 2025年第10期11卷

作者： Lee, Jun Yan Wu, Duo Guo, Xuanrui Tan, Jian Ding Yew, Teh Jia Ng, Zi Neng Bhuiyan, Mohammad Arif Sobhan Miraz, Mahdi H. School of Artificial Intelligence and Robotics Xiamen University Malaysia Selangor Malaysia School of Electronic Information and Electrical Engineering Shanghai Jiao Tong University Shanghai China School of Computing and Data Science Xiamen University Malaysia Selangor Malaysia School of Computing Faculty of Arts Science and Technology Wrexham University Wrexham United Kingdom Faculty of Computing Engineering and Science University of South Wales Pontypridd United Kingdom

The fifth-generation (5G) network faces limitations in supporting emerging applications, such as artificial intelligence (AI), virtual reality (VR) and digital twins. To overcome these confines, sub-Terahertz (sub-THz) and Terahertz (THz) technologies are considered to be key enablers of effective 6G wireless communications, offering higher transmission speeds, longer range and wider bandwidth. Achieving these capabilities requires careful engineering of 6G transceivers, with a focus on efficient power amplifiers (PAs) in the front-end, which play a critical role in effectively amplifying and transmitting signals over long distances. Complimentary metal-oxide-semiconductor (CMOS) technology-based PA in sub-THz suffers severe parasitic and limited maximum frequency, however, this has eventually been solved by different design architectures and scaling down of CMOS technology to break through the frequency limitations. In this article, we reviewed the potentials and capabilities of CMOS technology for designing 6G hardware, identified the state-of-art PA designs in the sub-THz band and then examined as well as compared the designs to identify the suitable design strategies for better performance. The circuit optimisation techniques, such as coupled-line, passive gain boosting method, zero-degree power splitting, load-pull matching, diode and capacitor linearisation for better gain, saturated output power and power added efficiency, are considered for the PA design architectures at different sub-THz bands. Furthermore, these methods are summarised and discussed with their advantages and disadvantages in lieu with their performances. The PA design trends, challenges and future perspectives are also presented and discussed. Therefore, this comprehensive review article will serve as a comparative study and reference for future PA designs for radio frequency integrated circuits (RFIC). © 2025 The Authors

关键词： 6G CMOS Front-end Power amplifier Sub-THz Transceiver Transmitter

来源：评论

学校读者我要写书评

暂无评论

Online learning of windmill time series using Long Short-term Cognitive Networks

arXiv

引用

arXiv 2021年

作者： Morales-Hernández, Alejandro Nápoles, Gonzalo Jastrzebska, Agnieszka Salgueiro, Yamisleydi Vanhoof, Koen Business Informatics Research Group Hasselt University Belgium Department of Cognitive Science & Artificial Intelligence Tilburg University Netherlands Faculty of Mathematics and Information Science Warsaw University of Technology Poland Department of Computer Science Faculty of Engineering Universidad de Talca Campus Curicó Chile

Forecasting windmill time series is often the basis of other processes such as anomaly detection, health monitoring, or maintenance scheduling. The amount of data generated by windmill farms makes online learning the most viable strategy to follow. Such settings require retraining the model each time a new batch of data is available. However, updating the model with new information is often very expensive when using traditional Recurrent Neural Networks (RNNs). In this paper, we use Long Short-term Cognitive Networks (LSTCNs) to forecast windmill time series in online settings. These recently introduced neural systems consist of chained Short-term Cognitive Network blocks, each processing a temporal data chunk. The learning algorithm of these blocks is based on a very fast, deterministic learning rule that makes LSTCNs suitable for online learning tasks. The numerical simulations using a case study involving four windmills showed that our approach reported the lowest forecasting errors with respect to a simple RNN, a Long Short-term Memory, a Gated Recurrent Unit, and a Hidden Markov Model. What is perhaps more important is that the LSTCN approach is significantly faster than these state-of-the-art models. © 2021, CC BY-NC-ND.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

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

请选择保存的检索档案：

请选择收藏分类：

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

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

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：