检索结果-内蒙古大学图书馆

Exploring 2D representation and transfer learning techniques for indoor localization

Multimedia Tools and Applications 2025年 1-13页

作者： Kerdjidj, O. Himeur, Y. Atalla, S. Copiaco, A. Sohail, S.S. Amira, A. Fadli, F. Mansoor, W. Gawanmeh, A. Centre of Development of Advanced Techniques Algiers Algeria College of Engineering and Information Technology University of Dubai Dubai United Arab Emirates Department of Computer Science University of Sharjah Sharjah United Arab Emirates Institute of Artificial Intelligence De Montfort University Leicester United Kingdom School of Computing Science and Engineering VIT Bhopal University Kothri Kalan India Department of Architecture & Urban Planning Qatar University Doha2713 Qatar

Indoor localization (IL) plays a critical role in various applications, including asset tracking, emergency response, and indoor navigation. Traditional methods relying on one-dimensional (1D) representations of received signal strength indicator (RSSI) data often suffer from limitations such as low accuracy, environmental sensitivity, and susceptibility to interference. To overcome these challenges, this paper proposes Scalogram-Based Transfer Learning for Indoor Localization (STL-IL), a novel framework that significantly enhances localization performance. The proposed approach transforms 1D RSSI signals into two-dimensional (2D) representations using scalogram-based time-frequency analysis, coupled with advanced transfer learning (TL) techniques. By leveraging the scalogram method to generate image representations of the data and employing diverse wavelet types. Furthermore, pre-trained deep learning models, including GoogleNet and SqueezeNet, are utilized to achieve state-of-the-art performance. Experimental results demonstrate that the STL-IL system achieves remarkable accuracy levels of 95% and 96% for SqueezeNet and GoogleNet, respectively. The proposed framework not only improves localization accuracy but also enhances robustness in dynamic and challenging environments, making it a promising solution for real-world indoor localization applications. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Deep learning

来源：评论

学校读者我要写书评

暂无评论

Integrating artificial neural networks and neighborhood analysis for enhanced urban growth prediction depending on planned future development areas: A case study of Ar-Ramtha, Jordan

引用

Journal of Urban Management 2025年

作者： Gharaibeh, Anne A. Al-Rabee, Mohammad S. Obeidat, Rasha Department of City Planning and Design College of Architecture and Design Jordan University of Science and Technology Irbid Jordan Department of Computer Science Faculty of Computer and Information Technology Jordan University of Science and Technology Irbid Jordan

Artificial intelligence (AI) is revolutionizing urban planning by enhancing the simulation of urban growth and land-use changes. Most AI models often fail to predict how proposed development areas will shape future expansion, leading to misaligned planning and urban sprawl. This study addresses these gaps by integrating master plans and spatial policies into an AI-driven urban growth model for Ar-Ramtha. A hybrid model combining artificial neural networks (ANN) and neighborhood analysis was developed to assess the influence of proposed development areas. Historical data (1990–2005) was used for training and validation, while 2005–2020 data was used for testing. Under-sampling techniques addressed data imbalance, and the PLADJ matrix measured urban growth patterns. Results indicate that incorporating neighborhood analysis and development areas improves model accuracy, achieving a validation accuracy of 92.9 %, an F1-micro of 91.0 %, and a Kappa of 82.1 %. Testing performance remained strong, with an F1-micro of 84.0 % and an accuracy of 91.2 %. Simulations suggest Ar-Ramtha is becoming more compact, with PLADJ values rising from 80.67 % in 2020 to 92.7 % by 2035. However, potential sprawl risks remain due to increased road paving and rising land prices. The study recommends defining development areas based on AI model predictions beyond current planning boundaries and adopting a ring road strategy to promote compact urban growth. These findings contribute to bridging gaps in AI-based urban modeling by improving the integration of planning policies, offering a robust tool for sustainable urban development. © 2025 The Authors

关键词： Artificial neural networks (ANN) Neighborhood analysis PLADJ Urban growth Urban sprawl

来源：评论

学校读者我要写书评

暂无评论

Maximizing Energy Efficiency in IRS-Assisted Phase Cooperative PS-SWIPT based Self-Sustainable IoT Network

IEEE Open Journal of the Communications Society

引用

IEEE Open Journal of the Communications Society 2025年 6卷 4311-4327页

作者： Sadia, Haleema Hassan, Ahmad Kamal Abbas, Ziaul Haq Abbas, Ghulam Baker, Thar Saeed, Nasir United Arab Emirates University College of Engineering Department of Electrical and Communication Engineering Al Ain United Arab Emirates Ghulam Ishaq Khan Institute of Engineering and Technology Faculty of Electrical Engineering Pakistan Ghulam Ishaq Khan Institute of Engineering and Technology Faculty of Computer Science and Engineering Pakistan University of Brighton School of Architecture Technology and Engineering Brighton BN2 4GJ United Kingdom

Power splitting based simultaneous wireless information and power transfer (PS-SWIPT) appears to be a promising solution to support future self-sustainable Internet of Things (SS-IoT) networks. However, the performance of these networks is constrained by radio frequency signal strength and channel impairments. To overcome this challenge, intelligent reflecting surfaces (IRSs) have recently been introduced in IoT networks to improve network efficiency by controlling signal reflections. In this article, we propose a novel IRS-enabled phase cooperative framework that aims to maximize the energy efficiency (EE) of the PS-SWIPT based SS-IoT network. Our framework leverages the idea of phase cooperation between two distinct networks without requiring additional hardware resources. We employed transmit beamforming (BF) at access points (APs) and phase shift optimization at the IRS with effective phase cooperation between APs. The EE maximization problem turns out to be NP-hard, so first, an alternating optimization (AO) is solved for the user’s network using heuristic BF approaches and phase optimization via semidefinite relaxation (SDR) approaches. To combat the computational complexity of AO, we also propose a low-complexity alternative solution by exploiting heuristic BF schemes and an iterative algorithm, i.e., the element-wise block-coordinate descent method for phase shifts optimization. The proposed solutions’ analysis shows low computational complexity and fast convergence, achieving near-optimal EE performance for different network settings. We conducted extensive simulations to evaluate our proposed framework’s performance and unveiled significant EE performance achieved by the IoT network consistent with the numerical findings. This work demonstrates the potential of IRS phase cooperation for enhancing the EE of different networks without constraining hardware resources. © 2020 IEEE.

关键词： beamforming energy efficiency Intelligent reflecting surface (IRS) Internet of Things (IoT) phase cooperative network power splitting SWIPT

来源：评论

学校读者我要写书评

暂无评论

FUCOM-Kendall Model for Evaluating E-Scooter Adoption: A Case Study in Catania, Italy

引用

Transportation Research Procedia 2025年 86卷 1-8页

作者： Vincenza Torrisi Rosalia Camporeale Sarbast Moslem Tiziana Campisi Department of Electric Electronic and Computer Engineering Catania Italy Department of Technology and Society Division of Transport and Roads Lund Sweden School of Architecture Planning and Environmental Policy University College Dublin Ireland Faculty of Engineering and Architecture University of Enna KORE Enna Italy

Electric scooters (e-scooters) have emerged as a fundamental component of urban micro-mobility, particularly in European cities during the Covid-19 pandemic. This study employs the Full Consistency Method (FUCOM) and Kendall rank correlation method to assess the preferences and priorities of university students in Catania, Italy, regarding e-scooter adoption. By analyzing influencing factors, this research offers a comprehensive understanding of student attitudes towards e-scooters and their adoption, contributing to strategies for enhancing sustainable urban mobility. The findings highlight significant preferences for factors such as parking ease, portability, and intermodality, while also identifying safety concerns and the need for better integration with public transport. These insights underscore the necessity for targeted improvements in e-scooter infrastructure and policies, aiming to optimize urban mobility and meet the specific needs of potential users. Further research will focus on comparative analysis across different urban settings, investigating addition user-centric factors to capture a more comprehensive view of e-scooter adoption.

关键词： e-scooter multi-criteria decision making FUCOM method Kendall rank correlation urban mobility

来源：评论

学校读者我要写书评

暂无评论

Optimal Synthesis of Memristive Mixed-Mode Circuits

Optimal Synthesis of Memristive Mixed-Mode Circuits

引用

Design, Automation and Test in Europe Conference and Exhibition

作者： Ilia Polian Xianyue Zhao Li-Wei Chen Felix Bayhurst Ziang Chen Heidemarie Schmidt Nan Du University of Stuttgart Institute of Computer Architecture and Computer Engineering Stuttgart Germany Department of Quantum Detection Friedrich Schiller University Jena Institute for Solid State Physics Leibniz Institute of Photonic Technology Jena Germany

ISBN: (数字)9783982674100

ISBN: (纸本)9798331534646

Memristive crossbars are attractive for in-memory computing due to their integration density combined with compute and storage capabilities of their basic devices. However, yield and fidelity of emerging memristive technologies can make their reliable operation unattainable, thus raising interest in simpler topologies. In this paper, we consider synthesis of Boolean functions on 1D memristive line arrays. We propose an optimal procedure that can fully utilize the rich electrical behavior of memristive devices, mixing stateful (resistance-input) and non-stateful (voltage-input) operations as desired by the designer, leveraging their respective strengths. The synthesis method is based on Boolean satisfiability (SAT) solving and supports flexible constraints to enforce, e.g., restrictions of the available peripher-als. We experimentally validate memristive logic circuits beyond individual logic gates by demonstrating the operation of a Galois field multiplier using a 1D line array of 10 memristors in parallel, highlighting the robust performance of our proposed mixed-mode circuit and its synthesis procedure.

关键词： Boolean functions Memristors Europe Logic gates In-memory computing Topology Logic Reliability Galois fields Logic arrays

来源：评论

学校读者我要写书评

暂无评论

SiamCircle: Trajectory Representation Learning in Free Settings 23rd

SiamCircle: Trajectory Representation Learning in Free Sett...

引用

23rd International Symposium on Intelligent Data Analysis, IDA 2025

作者： Nasri, Maedeh Baratchi, Mitra Koutamanis, Alexander Rieffe, Carolien Department of Developmental Psychology Leiden University Leiden Netherlands Leiden Institute of Advanced Computer Science Leiden University Leiden Netherlands Faculty of Architecture and the Built Environment Delft University of Technology Delft Netherlands Faculty of Electrical Engineering Mathematics and Computer Science University of Twente Enschede Netherlands Department of Psychology and Human Development Institute of Education UCL London United Kingdom

ISBN: (纸本)9783031913976

Trajectory representation learning (TRL) is an intermediate step in handling trajectory data to realize various downstream machine-learning tasks. While most previous TRL research focuses on modeling structured movements in large-scale urban spaces (e.g., cars or pedestrians on streets), this paper focuses on a more challenging scenario of modeling free movement in small-scale social spaces (e.g., children playing in a schoolyard). We present a TRL model, SiamCircle, to process raw trajectories without additional feature extraction to prevent information loss. SiamCircle adopts a Siamese network with Circle Loss to learn trajectory embeddings. Furthermore, SiamCircle employs a data augmentation process to enable self-supervised learning and enrich the input data to address the limited access to high-quality data and ground truth. We evaluate the performance of SiamCircle in downstream tasks using trajectory ranking and clustering performance via seven evaluation metrics collectively. Using an ablation study, we explored the impact of different loss functions on the model’s performance. Accordingly, we selected a 2-D convolutional design with Circle Loss as the best-performing model. In a comparative study, we compared our model against three other baselines. We observed up to 19% improvements in trajectory ranking tasks and achieved the highest average rank in supervised clustering tasks. © The Author(s), under exclusive license to Springer Nature Switzerland AG 2025.

关键词： Clustering Trajectory Representation Learning Triplet Loss

来源：评论

学校读者我要写书评

暂无评论

RGB-Based Category-Level Object Pose Estimation via Depth Recovery and Adaptive Refinement

引用

IEEE Robotics and Automation Letters 2025年第6期10卷 5377-5384页

作者： Yang, Hui Sun, Wei Liu, Jian Zheng, Jin Zeng, Zhiwen Mian, Ajmal Hunan University National Engineering Research Center for Robot Visual Perception and Control Technology College of Electrical and Information Engineering Changsha410082 China Hunan University State Key Laboratory of Advanced Design and Manufacturing for Vehicle Body Changsha410082 China Central South University School of Architecture and Art Changsha410082 China University of Western Australia Department of Computer Science and Software Engineering CrawleyWA6009 Australia

Category-level pose estimation methods have received widespread attention as they can be generalized to intra-class unseen objects. Although RGB-D-based category-level methods have made significant progress, reliance on depth image limits practical application. RGB-based methods offer a more practical and cost-effective solution. However, current RGB-based methods struggle with object geometry perception, leading to inaccurate pose estimation. We propose depth recovery and adaptive refinement for category-level object pose estimation from a single RGB image. We leverage DINOv2 to reconstruct the coarse scene-level depth from the input RGB image and propose an adaptive refinement network based on an encoder-decoder architecture to dynamically improve the predicted coarse depth and reduce its gap from the ground truth. We introduce a 2D–3D consistency loss to ensure correspondence between the point cloud obtained from depth projection and the objects in the 2D image. This consistency supervision enables the model to maintain alignment between the depth image and the point cloud. Finally, we extract features from the refined point cloud and feed them into two confidence-aware rotation regression branches and a translation and size prediction residual branch for end-to-end training. Decoupling the rotation matrix provides a more direct representation, which facilitates parameter optimization and gradient propagation. Extensive experiments on the REAL275 and CAMERA25 datasets demonstrate the superior performance of our method. Real-world estimation and robotic grasping experiments demonstrate our model robustness to occlusion, clutter environments, and low-textured objects. Our code and robotic grasping video are available at DA-Pose. © 2016 IEEE.

关键词： Depth perception

来源：评论

学校读者我要写书评

暂无评论

DHAG-DTA: Dynamic Hierarchical Affinity Graph Model for Drug-Target Binding Affinity Prediction

IEEE Transactions on Computational Biology and Bioinformatic...

引用

IEEE Transactions on Computational Biology and Bioinformatics 2025年第2期22卷 697-709页

作者： Cheng Wang Yang Liu Shitao Song Kun Cao Xiaoyan Liu Gaurav Sharma Maozu Guo School of Computer Science and Technology Harbin Institute of Technology Harbin China School of Electrical Engineering Liaoning University of Technology Liaoning China Department of Electrical and Computer Engineering University of Rochester Rochester NY USA School of Electrical and Information Engineering Beijing Key Laboratory of Intelligent Processing for Building Big Data Beijing University of Civil Engineering and Architecture Beijing China

Computational methods for predicting drug-target binding affinity (DTA) are critical for large-scale screening of prospective therapeutic compounds during drug discovery. Deep neural networks (DNNs) have recently shown significant promise for DTA prediction. By leveraging available data for training, DNNs can expand the use of DTA prediction to situations where only sequence information is available for potential drug molecules and their targets, and there is no prior knowledge regarding the molecular geometric conformations. We propose DHAG-DTA, a general dynamic hierarchical affinity graph DNN approach, for DTA prediction using molecular sequence information and already known drug-target interactions. DHAG-DTA introduces a two-level hierarchical graph structure: at the upper level, interactions between drug and target molecules are represented via an affinity graph and at the lower level, embedded molecular graphs represent interactions within the individual molecules. This allows for integration of information from both inter and intra molecular interactions for DTA prediction, which has also been addressed in other recent independent work. The fundamental innovations introduced by DHAG-DTA include: (a) a single overall hierarchical graph that allows better assimilation of information during the learning process compared with loosely-coupled individual graphs, (b) dynamic determination of the affinity graph structure via the introduction of unlabeled edges and a maximum entropy criterion for active edge selection, (c) skip connections in the DNN for fusing intra and inter molecular information, and (d) fusion of both model-based and similarity-based feature embeddings to get robust embeddings of unseen molecules. Experimental results on two common benchmark datasets demonstrate that DHAG-DTA outperforms other existing models on multiple evaluation metrics, achieving state-of-the-art performance.

关键词： Drugs Predictive models Graph neural networks Artificial neural networks Biological system modeling Proteins Atoms Computational modeling Bioinformatics Entropy

来源：评论

学校读者我要写书评

暂无评论

Customization of the text-to-image diffusion model by fine-tuning for the generation of synthetic images of cyanobacterial blooms in lentic water bodies

引用

Expert Systems with Applications 2025年 287卷

作者： Barrientos-Espillco, Fredy Pajares, Gonzalo López-Orozco, José A. Besada-Portas, Eva Department of Computer Architecture and Automation University Complutense of Madrid Madrid28040 Spain Institute for Knowledge Technology University Complutense of Madrid Madrid28040 Spain

Cyanobacterial blooms emerge unpredictably on the surface of lentic water bodies, posing both ecological threats and public health risks. To effectively monitor these events, this study introduces the use of Machine Vision Systems (MVS) integrated into Autonomous Surface Vehicles (ASVs). These ASVs are capable of autonomous and safe navigation, enabling them to detect cyanobacterial blooms while avoiding obstacles. Convolutional Neural Networks (CNNs) are employed for early detection and continuous monitoring, but their effectiveness hinges on access to large, high-quality training datasets. Due to the sporadic and uncontrollable nature of bloom occurrences, acquiring sufficient real-world images for training and validating CNN models is a significant challenge. To overcome this, the Stable Diffusion XL (SDXL) text-to-image generative model is utilized to produce realistic synthetic images, ensuring a sufficient dataset for training. However, SDXL alone struggles to accurately depict cyanobacterial blooms. To address this limitation, DreamBooth is used to fine-tune SDXL with a small set of real bloom-specific image patches. To ensure the diversity of the synthetic dataset, detailed prompts for SDXL are generated using a Large Language Model (LLM). The combination of SDXL fine-tuning with LLM-driven prompts design applied to environmental monitoring and autonomous navigation in lentic environments represents the core innovation of this work. A dual-task CNN model is then trained on the synthetic dataset to simultaneously detect blooms and obstacles. Experimental results demonstrate the effectiveness and novelty of the proposed approach, showing improvements of up to 15.74% in object detection and 6.48% in semantic segmentation compared to the baseline dataset. © 2025 The Author(s)

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Integrating Structural Information: Comparing Classification-Based and Segmentation-Based Predictors in Bioinformatics (Case Study: Protein-Peptide Region Residue-Level Interaction) Using Structural Information

Integrating Structural Information: Comparing Classification...

引用

International CSI computer Conference, CSICC

作者： Shima Shafiee Abdolhossein Fathi Faezeh Safari Department of Computer Engineering and Information Technology Razi University Kermanshah Iran Faculty of Eng. Architecture and Info. Tech. School of Electrical Eng. and Computer Science University of Queensland Brisbane QLD Australia

ISBN: (数字)9798331523114

ISBN: (纸本)9798331523121

Understanding the regions where proteins bind to peptides is vital for studying diseases like cancer, as it sheds light on various cellular processes and aids in drug discovery. Although researchers can experimentally investigate protein-peptide interactions at the region level by analyzing the structures of their complexes, only a limited number of these structures are known. Additionally, experiments to determine these interactions can be both costly and time-consuming. Computational methods provide an alternative and complementary approach to experimental techniques. As a result, using computational methods to predict peptide-binding regions can significantly enhance the efficiency and cost-effectiveness of experimental research. To leverage these advantages, this study introduces a new method named Structure-based Prediction of Residue-level Interaction of Peptides (SPRINT-Pep), designed to pinpoint the binding regions between proteins and peptides. This approach leverages two distinct predictive models: a classification-based model combining DeepResNet with a Support Vector Machine (SVM) and a segmentation-based model called DeepU-Net. Our predictions outperform other state-of-the-art methods in terms of F-measures (by at least 23%), precision (by at least 3.7%), and the balance between sensitivity and specificity (by about 7.7%). These findings demonstrate the proficiency of the proposed method.

关键词： Proteins Support vector machines Deep learning Peptides Predictive models Sensitivity and specificity Prediction algorithms Robustness Drug discovery Testing

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：