检索结果-内蒙古大学图书馆

Artificial Intelligence Science and engineering 2025年第1期1卷 64-78页

作者： Shanshan ZHAO Linhao ZHAO Shiping WEN Long CHENG Faculty of Engineering and Information Technology Australian AI Institute The University of Technology Sydney Sydney Australia State Key Laboratory of Multimodal Artificial Intelligence Systems Chinese Academy of Sciences Institute of Automation Beijing China School of Artificial Intelligence University of Chinese Academy of Sciences Beijing China

This paper explores the issue of secure synchronization control in piecewise-homogeneous Markovian jump delay neural networks affected by denial-of-service (DoS) attacks. Initially, a novel memory-based adaptive event-triggered mechanism (MBAETM) is designed based on sequential growth rates, focusing on event-triggered conditions and thresholds. Subsequently, from the perspective of defenders, non-periodic DoS attacks are re-characterized, and a model of irregular DoS attacks with cyclic fluctuations within time series is further introduced to enhance the system's defense capabilities more effectively. Additionally, considering the unified demands of network security and communication efficiency, a resilient memory-based adaptive event-triggered mechanism (RMBAETM) is proposed. A unified Lyapunov-Krasovskii functional is then constructed, incorporating a loop functional to thoroughly consider information at trigger moments. The master-slave system achieves synchronization through the application of linear matrix inequality techniques. Finally, the proposed methods' effectiveness and superiority are confirmed through four numerical simulation examples.

关键词： Adaptive systems Event detection Neural networks Time series analysis Network security Denial-of-service attack Numerical simulation Main-secondary Delays Synchronization

来源：评论

学校读者我要写书评

暂无评论

Advanced AI techniques for root disease classification in dental X-rays using deep learning and metaheuristic approach

引用

Intelligent Systems with Applications 2025年 26卷

作者： Enkvetchakul, Prem Khonjun, Surajet Pitakaso, Rapeepan Srichok, Thanatkij Luesak, Peerawat Kaewta, Chutchai Gonwirat, Sarayut Boonmee, Chawis Noowattana, Matus Srisuwandee, Thitinon Department of Information Technology Faculty of Science Buriram Rajabhat University Buriram31000 Thailand Artificial Intelligence Optimization SMART Laboratory Industrial Engineering Department Faculty of Engineering Ubon Ratchathani University Ubon Ratchathani34190 Thailand Department of Industrial Engineering Faculty of Engineering Rajamangala University of Technology Lanna Chiang Rai57120 Thailand Digital Innovation Faculty of Computer Science Ubon Ratchathani Rajabhat University Mueang Ubon Ratchathani34000 Thailand Department of Computer Engineering and Automation Kalasin University Kalasin46000 Thailand Department of Industrial Engineering Faculty of Engineering Chiang Mai University Chiang Mai Thailand Dental Department Samrong Hospital Samrong District Ubon Ratchathani34360 Thailand

Root dental diseases remain among the most diagnostically challenging conditions in oral healthcare, often leading to treatment delays and suboptimal outcomes. This study is motivated by the limitations of existing automated diagnostic systems, which tend to focus on superficial abnormalities and overlook complex root pathologies such as pulpal infections, periapical lesions, and progressive periodontitis. To bridge this critical gap, we propose an advanced AI-based classification model that integrates ensemble deep learning architectures with a hybrid metaheuristic optimization strategy-namely, the non-population-based Artificial Multiple Intelligence System (np-AMIS) for image augmentation and the population-based AMIS (pop-AMIS) for adaptive decision fusion. This dual-phase approach enhances feature diversity, classification robustness, and computational efficiency. The model was trained and validated on two proprietary datasets, TD-1 and TD-2, achieving classification accuracies of 98.87 % and 98.41 %, respectively. It was further implemented in a real-world application via the Automated Teeth Disease and Abnormality Classification System (A-TD-A-CS), demonstrating 98.95 % accuracy, a rapid response time of 1.5 s, and a System Usability Scale (SUS) score of 94.5 from dental professionals. The system's ability to accurately identify multiple root disease categories highlights its clinical viability and transformative potential. In addition to its current performance, this study lays the groundwork for future extensions to multi-center datasets and cross-modality diagnostics using cone-beam CT or intraoral scans, further advancing intelligent dental care. © 2025

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

TPCap: Unlocking Zero-Shot Image Captioning with Trigger-Augmented and Multi-Modal Purification Modules

arXiv

引用

arXiv 2025年

作者： Zhang, Ruoyu Wang, Lulu He, Yi Pan, Tongling Yu, Zhengtao Li, Yingna Faculty of Information Engineering and Automation Kunming University of Science and Technology Kunming650504 China Key Laboratory of Computer Technologies Application Kunming650504 China Hongyun Honghe Group Honghe Cigarette Factory Kunming650202 China

Recent advancements in large language models (LLMs) have significantly enhanced the fluency and logical coherence of image captioning. Retrieval-Augmented Generation (RAG) is widely adopted to incorporate external knowledge into LLMs;however, existing RAG-based methods rely on separate retrieval banks, introducing computational overhead and limiting the utilization of LLMs’ inherent zero-shot capabilities. To address these limitations, we propose TPCap, a novel trigger-augmented and multi-modal purification framework for zero-shot image captioning without external retrieval libraries. TPCap consists of two key components: trigger-augmented (TA) generation and multi-modal purification (MP). The TA module employs a trigger projector with frozen and learnable projections to activate LLMs’ contextual reasoning, enhance visual-textual alignment, and mitigate data bias. The MP module further refines the generated entity-related information by filtering noise and enhancing feature quality, ensuring more precise and factually consistent captions. We evaluate TPCap on COCO, NoCaps, Flickr30k, and WHOOPS datasets. With only 0.82M trainable parameters and training on a single NVIDIA RTX 4090 GPU, TPCap achieves competitive performance comparable to state-of-the-art models. Copyright © 2025, The Authors. All rights reserved.

关键词： information filtering

来源：评论

学校读者我要写书评

暂无评论

Study of structural, electronic, phonon, thermodynamic, and hydrogen storage properties of hydride Rb2AsSnH6 perovskites: DFT insights

引用

International Journal of Hydrogen Energy 2025年 133卷 225-234页

作者： Zang, Zhaoliang Lu, Xu Hussain, Asghar Shahzad, Muhammad Khuram Hussain, Shoukat Kumar, Abhinav Rekha, M.M. Pattanayak, Binayak Tirth, Vineet Hussien, Mohamed Mechanical Engineering and Automation College of Logistics Engineering Shanghai Maritime University 1550 Haigang Avenue Lingang New City Pudong New Area Shanghai201306 China Safety Engineering College of Marine Science and Engineering Shanghai Maritime University 1550 Haigang Avenue Lingang New City Pudong New Area Shanghai201306 China Institute of Physics Khwaja Fareed University of Engineering and Information Technology Rahim Yar Khan64200 Pakistan Department of Nuclear and Renewable Energy Ural Federal University Named after the First President of Russia Boris Yeltsin Ekaterinburg620002 Russia Centre for Research Impact & Outcome Chitkara University Institute of Engineering and Technology Chitkara University Punjab Rajpura140401 Azerbaijan Department of Mechanical Engineering and Renewable Energy Technical Engineering College The Islamic University Najaf Iraq Karnataka Bangalore India Odisha Bhubaneswar751030 India Central Labs King Khalid University AlQura'a P.O. Box 960 Abha Saudi Arabia Mechanical Engineering Department College of Engineering King Khalid University Abha61421 Saudi Arabia Department of Chemistry Faculty of Science King Khalid University P.O. Box 9004 Abha61413 Saudi Arabia

In recent years, scientists have concentrated extensively on hydrogen storage materials due to their potential to store massive amounts of clean energy. In this study, the structural, mechanical, and optoelectronic properties of the novel perovskite hydride Rb2AsSnH6 are computationally explored using the CASTEP package with the GGA-PBE functional to evaluate its appropriateness for hydrogen storage applications. The optimized cubic crystal structure of Rb2AsSnH6 exhibits a lattice constant of 8.50 Å. The material demonstrates thermodynamic and mechanical stability, as evidenced by its positive phonon frequencies, positive elastic constants (meeting Born stability criteria), and negative formation enthalpy. Electronic band structure and density of states (DOS) analyses confirm its metallic nature, with overlapping valence and conduction bands yielding a zero bandgap. Mechanical properties, including Pugh's ratio (0.43), Poisson's ratio (2.21), and Cauchy's pressure (C11−C44 = 6.72 GPa), suggest ionic bonding and ductile behavior. Optical properties such as dielectric function, absorption spectra, refractive index, and conductivity further indicate that Rb2AsSnH6 is promising for optoelectronic devices, photonics, and energy storage systems. Finally, the calculated gravimetric hydrogen storage capacity of 1.63 wt % highlights its potential for practical hydrogen storage applications. These findings suggest that Rb2AsSnH6 is a compelling candidate for advancing hydrogen storage technologies and multifunctional material design. © 2025 Hydrogen Energy Publications LLC

关键词： Temperature distribution

来源：评论

学校读者我要写书评

暂无评论

STAR: Screen Time and Actor Recognition in Video Content

STAR: Screen Time and Actor Recognition in Video Content

引用

46th Annual Conference of the German Association for Pattern Recognition, DAGM-GCPR 2024

作者： Kerepecky, Tomas Sroubek, Filip Zitova, Barbara Flusser, Jan Institute of Information Theory and Automation The Czech Academy of Sciences Pod Vodarenskou vezi 4 Prague18200 Czech Republic Faculty of Nuclear Sciences and Physical Engineering CTU in Prague Brehova 78/7 Prague11519 Czech Republic

ISBN: (纸本)9783031851865

Accurately measuring the duration of actors’ presence in videos is a challenging task that goes beyond actor recognition. We propose the STAR pipeline, the new model designed to analyze the time performers appear on screen across diverse video content, including movies and TV shows. The proposed model has been successfully deployed and tested by the Czech TV infrastructure provider. Our pipeline uses machine learning techniques for shot detection, face detection, tracking, recognition, and introduces a novel shot-based method for calculating screen time. We present extensive experiments proving the robustness and real-time performance of our approach. Alongside the pipeline, we introduce the STAR dataset to address the need for high-quality benchmarks in evaluating screen time models, now available for download. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

PiVoD: Pitch, Volume and Duration Invariant Representation for Music Tone Quality Evaluation

IEEE Transactions on Audio, Speech and Language Processing

引用

IEEE Transactions on Audio, Speech and Language Processing 2025年 33卷 613-626页

作者： Yixin Wang Xiaohong Guan Youtian Du Chenxu Wang Xiaobing Li Yu Pan Center for Art & Science and Presentation & Communication Frontier Institute of Science and Technology Xi'an Jiaotong University Xi'an China MOE KLINNS Laboratory Faculty of Electronics and Information Engineering Xi'an Jiaotong University Xi'an China Department of Automation Center for Intelligent and Networked Systems Tsinghua University Beijing China Department of AI Music and Music Information Technology Central Conservatory of Music Beijing China

Tone quality is of pivotal importance in the auditory perception of musical performance. Depending on the performer and the instrument, tone quality evaluation is subjective and time-consuming, with inherent difficulties stemming from the absence of precise measurement methods. In this study, we develop a novel method for tone quality evaluation utilizing an adversarial domain-invariant learning strategy to construct a representation invariant to changes in pitch, volume, and duration. The wide-band Mel frequency cepstral coefficients are employed for pitch-invariant feature extraction and instance normalization for volume invariance. An adversarial-trained time-delay neural network encoder is developed for enhancing pitch and duration invariance via random pitch shift and temporal segmentation. Experiments conducted on our curated dataset and the Good-sound dataset show that significant improvements from the new method are achieved in evaluating tone quality ascribed to performers and instruments, yielding a 15.3% and 9.5% increase in classification accuracy, respectively, compared to classical feature-based techniques. Remarkably, the class-wise outcomes exhibit enhancements in F-scores of 33.6% and 9.8% for each respective dataset. Ablation studies on pitch, volume, and duration invariance further underscore the efficacy of our approach. This substantial enhancement over existing methods presents a novel perspective on tone quality representation and offers a practical resource for music performance analysis.

关键词： Instruments Music Feature extraction Accuracy Speech processing Timbre Harmonic analysis Deep learning Training Power harmonic filters

来源：评论

学校读者我要写书评

暂无评论

LCS4: Linear Complexity State-Space Transformers with Multi-Scale Overlapping Scans for Efficient Semantic Segmentation

引用

IEEE Transactions on Instrumentation and Measurement 2025年 74卷

作者： Wang, Yueying Qing, Yuhao Fang, Zhijun Zhao, Xudong Shanghai University School of Mechatronic Engineering and Automation Shanghai200444 China Donghua University School of Computer Science and Technology Shanghai200051 China Dalian University of Technology Faculty of Electronic Information and Electrical Engineering Dalian116024 China

With the advancement of intelligent systems, semantic segmentation—an essential aspect of intelligent perception—is garnering significant research interest. However, convolutional neural network (CNN)-based methods exhibit limitations in processing long-range features. Although transformer-based models capably handle global information, their application in real-time semantic segmentation is computationally challenging due to their quadratic complexity. To address these issues, we introduce a state space model-based transformer architecture, LCS4, for semantic segmentation. LCS4 integrates the strengths of the Mamba system with the transformer architecture, enabling the modeling of remote contextual information with linear complexity. Additionally, the newly proposed multi-scale spatial overlap scanning module selectively processes spatial features from multiple directions, enhancing the construction of robust hierarchical features that improve both spatial and semantic continuity. Extensive experiments conducted across various datasets confirm the efficacy of LCS4 in semantic segmentation tasks, demonstrating lower parameter counts and higher accuracy. Specifically, on the Cityscapes dataset, LCS4 achieves a mean Intersection Over Union (mIOU) of 79.1 with only 2.92 million parameters and 11.8 GFLOPS. In comparison with WaSR on the MODD dataset, LCS4 demonstrates competitive performance with merely 6% of the parameters and operates approximately six times faster. © 1963-2012 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

Exploring Embodied Emotional Communication: A Human-oriented Review of Mediated Social Touch

arXiv

引用

arXiv 2025年

作者： He, Liwen Guo, Zichun Mo, Yanru Wen, Yve Wang, Yun School of Mechanical Engineering and Automation Beihang University Beijing100191 China College of Arts and Design Beijing University of Chemical Technology Beijing100029 China Faculty of Industrial Design Engineering Delft University of Technology Delft10587 Netherlands School of electronic information and electrical engineering Shanghai Jiao Tong University Shanghai200240 China

This paper offers a structured understanding of mediated social touch (MST) using a human-oriented approach, through an extensive review of literature spanning tactile interfaces, emotional information, mapping mechanisms, and the dynamics of human-human and human-robot interactions. By investigating the existing and exploratory mapping strategies of the 37 selected MST cases, we established the emotional expression space of MSTs that accommodated a diverse spectrum of emotions by integrating the categorical and Valence-Arousal model, showcasing how emotional cues can be translated into tactile signals. Based on the expressive capacity of MSTs, a practical design space was structured encompassing factors such as the body locations, device form, tactile modalities and parameters. We also proposed various design strategies for MSTs including workflow, evaluation methods, and ethical and cultural considerations, as well as several future research directions. MSTs’ potential is reflected not only in conveying emotional information but also in fostering empathy, comfort, and connection in both human-human and human-robot interactions. This paper aims to serve as a comprehensive reference for design researchers and practitioners, which helps expand the scope of emotional communication of MSTs, facilitating the exploration of diverse applications of affective haptics, and enhancing the naturalness and sociability of haptic *** Codes 68T40 (Primary), 92B20 (Secondary), 68U05 (Secondary) Copyright © 2025, The Authors. All rights reserved.

关键词： Mapping

来源：评论

学校读者我要写书评

暂无评论

Multimodal Image Matching based on Frequency-domain information of Local Energy Response

arXiv

引用

arXiv 2025年

作者： Yang, Meng Chen, Jun Gong, Wenping Wei, Longsheng Tian, Xin School of Automation China University of Geosciences Wuhan430074 China Hubei Key Laboratory of Advanced Control and Intelligent Automation for Complex Systems Wuhan430074 China Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan430074 China Faculty of Engineering China University of Geosciences Wuhan430074 China Electronic Information School Wuhan University Wuhan430072 China

Complicated nonlinear intensity differences, nonlinear local geometric distortions, noises and rotation transformation are main challenges in multimodal image matching. In order to solve these problems, we propose a method based on Frequency-domain information of Local Energy Response called FILER. The core of FILER is the local energy response model based on frequency-domain information, which can overcome the effect of nonlinear intensity differences. To improve the robustness to local nonlinear geometric distortions and noises, we design a new edge structure enhanced feature detector and convolutional feature weighted descriptor, respectively. In addition, FILER overcomes the sensitivity of the frequency-domain information to the rotation angle and achieves rotation invariance. Extensive experiments multimodal image pairs show that FILER outperforms other state-of-the-art algorithms and has good robustness and *** Codes 68U10 (Primary), 68T45, 68Wxx (Secondary) Copyright © 2025, The Authors. All rights reserved.

关键词： Mathematical transformations

来源：评论

学校读者我要写书评

暂无评论

Dual-task Mutual Reinforcing Embedded Joint Video Paragraph Retrieval and Grounding

引用

IEEE Transactions on Multimedia 2025年

作者： Wang, Mengzhao Li, Huafeng Zhang, Yafei Li, Jinxing Xie, Minghong Tao, Dapeng Kunming University of Science and Technology Faculty of Information Engineering and Automation Kunming650500 China Harbin Institute of Technology at Shenzhen Shenzhen518055 China Shenzhen Key Laboratory of Visual Object Detection and Recognition Shenzhen518055 China Yunnan University FIST LAB School of Information Science and Engineering Kunming650091 China

Video Paragraph Grounding (VPG) aims to precisely locate the most appropriate moments within a video that are relevant to a given textual paragraph query. However, existing methods typically rely on large-scale annotated temporal labels and assume that the correspondence between videos and paragraphs is known. This is impractical in real-world applications, as constructing temporal labels requires significant labor costs, and the correspondence is often unknown. To address this issue, we propose a Dual-task Mutual Reinforcing Embedded Joint Video Paragraph Retrieval and Grounding method (DMR-JRG). In this method, retrieval and grounding tasks are mutually reinforced rather than being treated as separate issues. DMR-JRG mainly consists of two branches: a retrieval branch and a grounding branch. The retrieval branch uses inter-video contrastive learning to roughly align the global features of paragraphs and videos, reducing modality differences and constructing a coarse-grained feature space to break free from the need for correspondence between paragraphs and videos. Additionally, this coarse-grained feature space further facilitates the grounding branch in extracting fine-grained contextual representations. In the grounding branch, we achieve precise cross-modal matching and grounding by exploring the consistency between local, global, and temporal dimensions of video segments and textual paragraphs. By synergizing these dimensions, we construct a fine-grained feature space for video and textual features, greatly reducing the need for large-scale annotated temporal labels. Meanwhile, we design a grounding reinforcement retrieval module (GRRM) that brings the coarse-grained feature space of the retrieval branch closer to the fine-grained feature space of the grounding branch, thereby reinforcing retrieval branch through grounding branch, and finally achieving mutual reinforcement between tasks. Extensive experiments on three challenging datasets demonstrate the effec

关键词： Electric grounding

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：