检索结果-内蒙古大学图书馆

Eddy Detection of Lakes in the Qinghai–Tibet Plateau Based on Optical Remote Sensing and Deep Learning

IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing 2025年 18卷 11996-12007页

作者： Bai, Bingxin Mu, Lixia Yang, Jie Huang, Baoxiang Ma, Chunyong Chen, Ge Ocean University of China Faculty of Information Science and Engineering Qingdao266100 China Laoshan Laboratory Qingdao266237 China Ocean University of China College of Marine Technology Qingdao266100 China Qingdao University College of Computer Science and Technology Qingdao266071 China

Lake eddies are dynamic phenomena prevalent in large lake systems, playing a critical role in affecting lake physics, nutrient transport, and ecological balance. Efficient and precise detection of these features is essential for advancing scientific understanding of lake dynamics and for improving environment management strategies. Despite their importance, the study of lake eddies remains under-explored, and current detection methods are insufficient. This study addresses this gap by proposing a lake eddy identification framework based on the YOLOv7 deep learning model, utilizing Landsat 8/9 satellite imagery acquired from April 2013 to December 2023. Qinghai Lake, Selin Co, and Nam Co on the Qinghai–Tibet Plateau were selected as the study areas. The method involves the development of an image enhancement process to optimize the visibility of lake eddies in remote sensing imagery, followed by the implementation of the YOLOv7 model for detection. The framework identified 912 cyclonic eddies and 102 anticyclonic eddies on the three aforementioned lakes, providing detailed information on their spatial distribution and morphology. This approach demonstrates significant potential for advancing the study of lake hydrodynamics and energy transport substances in aquatic ecosystems. © 2008-2012 IEEE.

关键词： Aquatic ecosystems

来源：评论

学校读者我要写书评

暂无评论

Temporal Dynamics of Consumer Engagement in E-Commerce

Temporal Dynamics of Consumer Engagement in E-Commerce

引用

2025 International Conference on Computing for Sustainability and Intelligent Future, COMP-SIF 2025

作者： Thiyagarajan, Gomathi Swathi, Y. Cmr Institute of Technology Department of Computer Applications Bengaluru India Gopalan College of Engineering and Management Department of Computer Science and Engineering Bengaluru India

ISBN: (纸本)9798331538538

This paper explores into an indepth analysis of consumer behavior on an e-commerce platform, extending the study period to include November 2019 through February 2020. Looking at user interactions that come under the categories of views, carts, purchases, and removals from cart along with process discovery visualizations will help identify complex patterns of engagement, measure the impact of interface changes and quantify marketing intervention's successes. This research uses sophisticated data analytics through process mining to provide recommendations for improving the e-commerce journey through increased user satisfaction and better conversion rates. © 2025 IEEE.

关键词： Marketplaces

来源：评论

学校读者我要写书评

暂无评论

Frequency Attacks Based on Invertible Neural Networks

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2025年第2期6卷 292-300页

作者： Shao, Ming-Wen Yang, Jian-Xin Meng, Ling-Zhuang Hu, Zhi-Yong China University of Petroleum School of Computer Science and Technology East China Qingdao266580 China China University of Petroleum College of Control Science and Engineering East China Qingdao266580 China

Adversarial attacks reveal the vulnerability of classifiers based on deep neural networks to well-designed perturbations. Most existing attack methods focus on adding perturbations directly to the pixel space. However, the perturbations generated by these methods may be easily perceived by humans. To alleviate the aforementioned problem, we propose a novel high-frequency attack based on invertible neural networks (HA-INN) that relies on INNs to adds perturbations to the high-frequency space of the image instead of the pixel space. In this way, we can fool the classifier while the perturbations are not easily detected by humans. Specifically, we introduce INNs to separate the high-frequency and low-frequency components of the image. And the low-frequency components are guaranteed to be reconstructed as the original image, while the high-frequency components are replaced with resampled high-frequency latent variables with additional adversarial information. Then, the low-frequency components and the high-frequency components with adversarial information are inversely fed into the INN to generate effective adversarial examples. Extensive experiments on two datasets (CIFAR-10 and CIFAR-100) show that our method can generate misleading and transferable cross-architectural adversarial examples with greatly reduced computational resource requirements. Under the white-box setting, the attack success rate of CIFAR10 and CIFAR100 is 99.8% and 99.74%, respectively. Further, under the black-box setting, the adversarial examples generated by our method are more effective than the other methods. © 2020 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing waste sorting and recycling efficiency: robust deep learning-based approach for classification and detection

引用

Neural Computing and Applications 2025年第6期37卷 4567-4583页

作者： Sayem, Faizul Rakib Islam, Md. Sakib Bin Naznine, Mansura Nashbat, Mohammad Hasan-Zia, Mazhar Kunju, Ali K Ansaruddin Khandakar, Amith Ashraf, Azad Majid, Molla Ehsanul Kashem, Saad Bin Abul Chowdhury, Muhammad E. H. Department of Electrical and Electronic Engineering University of Dhaka Dhaka Bangladesh Department of Biomedical Engineering Military Institute of Science and Technology Dhaka Bangladesh Department of Computer Science & Engineering Rajshahi University of Engineering & Technology Rajshahi6204 Bangladesh Chemical Engineering Department University of Doha for Science and Technology Doha Qatar Department of Electrical Engineering Qatar University Doha2713 Qatar Computer Applications Department Academic Bridge Program Qatar Foundation Doha Qatar Department of Computing Science AFG College With the University of Aberdeen Doha Qatar

Given the severity of waste pollution as a major environmental concern, intelligent and sustainable waste management is becoming increasingly crucial in both developed and developing countries. The material composition and volume of urban solid waste are key considerations in processing, managing, and utilizing city waste. Deep learning technologies have emerged as viable solutions to address waste management issues by reducing labor costs and automating complex tasks. However, the limited number of trash image categories and the inadequacy of existing datasets have constrained the proper evaluation of machine learning model performance across a large number of waste classes. In this paper, we present robust waste image classification and object detection studies using deep learning models, utilizing 28 distinct recyclable categories of waste images comprising a total of 10,406 images. For the waste classification task, we proposed a novel dual-stream network that outperformed several state-of-the-art models, achieving an overall classification accuracy of 83.11%. Additionally, we introduced the GELAN-E (generalized efficient layer aggregation network) model for waste object detection tasks, obtaining a mean average precision (mAP50) of 63%, surpassing other state-of-the-art detection models. These advancements demonstrate significant progress in the field of intelligent waste management, paving the way for more efficient and effective solutions. © The Author(s) 2024.

关键词： Deep learning for waste management Intelligent waste management Object detection in waste management Sustainable waste processing Urban solid waste Waste image classification

来源：评论

学校读者我要写书评

暂无评论

Automobile Entry System Based on Face Recognition

Journal of Network Intelligence

引用

Journal of Network Intelligence 2025年第1期10卷 573-585页

作者： Fang, Weidong Yu, Zuoxi Pan, Jeng-Shyang School of Electronic Electrical Engineering and Physics Fujian Provincial Key Laboratory of Automotive Electronics and Electric Drive Fujian University of Technology Fuzhou350118 China College of Computer Science and Engineering Shandong University of Science and Technology Qingdao266590 China Chaoyang University of Technology Taichung413310 Taiwan

With the development of technology, the automobile has become an indispensable part of people’s daily lives. People’s needs for automobile entry systems have also changed, in automobile safety and ease of use have become more and more important. In recent years, face recognition technology has made significant progress, and face recognition technology has been widely used in various fields, especially face recognition based on deep learning has great advantages in accuracy, recognition speed, and security, and can provide a more secure and reliable way of identity verification. Traditional automobile entry systems usually use mechanical keys and remote control keys, which do not remove the key and have certain shortcomings in security and user experience. Face recognition-based car entry systems can make up for these shortcomings and provide a more convenient and intuitive user experience. Combining face recognition with automobiles is also a hot topic in current scientific research. In this paper, a set of automobile entry systems with high efficiency and security is designed according to the face recognition method research, using three deep learning models: face detection, live body detection, and face recognition. In face detection, the RetinaFace lightweight model is used the network structure is improved, and the detection speed is increased by 14.9%. For face live detection and face recognition, the MobileFaceNet lightweight network is used as the base network for live detection and face recognition, achieving a 98.9% accuracy rate on the CelebA Spoof live detection dataset. In face recognition, feature extraction is performed on the detected faces after face detection, and the recognition results are output by comparing with the recorded faces. Improvements to its network improved the recognition accuracy by 0.18%, 0.77%, and 0.73% on the LFW, CFP FP, and AgeDB30 datasets, respectively. The model was deployed on Raspberry Pi and connected to CANoe via CAN bu

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Bidirectional Authentication Protocol for RFID Systems Based on Mms

引用

International Journal of Network Security 2025年第1期27卷 72-80页

作者： He, Xiao-ping Liu, Dao-wei Wu, Yang School of Data and Computer Science Guangdong Peizheng College Guangzhou510800 China Engineering College Guangzhou College of Technology and Business Guangzhou510006 China College of Information and Communication National University of Defense Technology Wuhan430010 China

Data exchange between electronic tags and the reader-writer in RFID systems is based on the wireless channel. Due to the inherent openness of the wireless channel, the transmitted data information is easy to be eavesdropped by other illegal users. In order to ensure data security, an ultra-lightweight mutual authentication protocol is proposed to solve this problem. In this paper, we independently design a new encryption algorithm, namely the product module synthesis operation. This algorithm is implemented based on bit operations, which can greatly reduce the total computation burden. Using the inherent Hamming weight and anti-Hamming weight information of encrypted parameters, different synthesis operations are carried out based on their product size relationship. Through the protocol analysis, it is shown that the proposed protocol has high security performance and meets the current computational requirements of low-cost RFID systems. © (2025), (Femto Technique Co., Ltd.). All rights reserved.

关键词： Radio frequency identification (RFID)

来源：评论

学校读者我要写书评

暂无评论

A novel workload forecasting model for cloud computing using ALAA-DBN algorithm

引用

Multimedia Tools and Applications 2025年第13期84卷 11383-11407页

作者： Karimunnisa, Syed Gopu, Arunkumar Rao, T. Prabhakara Ayyadurai, M. Kumar, ESaravana Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation AP VaddeswaramGuntur-522501 India School of Computer Science and Engineering VIT - AP University Amaravati522237 India Department of CSE Godavari Institute Of Engineering And Technology AP Rajahmundry533296 India Department of ECE SRM Institute of Science and Technology Ramapuram Campus Chennai India Department of CSE The Oxford College of Engineering Bengaluru India

Cloud Computing (CC) generally exhibits varying workload patterns. This autoscaling feature of CC has been extensively managed through predictive cloud resource management approaches. For this reason, a solitary forecasting model is not sufficient to forecast various workload patterns of CC web applications. With the intention of developing a proficient CC model, this paper implements a novel cloud workload forecasting model with the efficiency of DL algorithms like Deep Belief Network (DBN) and Lion Algorithm (LA). Here, the previous workload models are assessed and trained via DBN, and the prediction efficacy is enhanced by optimizing the DBN's hidden neurons through LA. The proposed workload forecast model for the cloud is accomplished by tuning the significant parameters via the Adaptive Lion Algorithm Assisted-DBN (ALAA-DBN)model. Besides, a well-organized resource provisioning pattern is followed to manage the under/over provisioning issues. The proposed CC model also considers providing better Quality of Service (QoS) to the users without violating the Service Level Agreement (SLA). Eventually, the significance of the proposed ALAA-DBN is verified through a comparative analysis with state-of-the-art models. © The Author(s), under exclusive licence to Springer science+Business Media, LLC, part of Springer Nature 2024.

关键词： Quality of service

来源：评论

学校读者我要写书评

暂无评论

Improvised Spectral Efficiency and Channel Estimation Parameters in Visible Light Vehicular Communication by Integrating Simulation of Urban Mobility Data

引用

IEEE Access 2025年 13卷 70828-70848页

作者： Wheeb, Ali H. Maurya, Sudhanshu Karimullah, Shaik Shaik, Fahimuddin Chowdhury, Subrata Siva Prasad, B.V.V. Sinha, Anurag Sood, Adamya Hussain, Syed Abid University of Baghdad College of Engineering Department of Aeronautical Engineering Baghdad10071 Iraq Department of Computer Science and Engineering Pune412115 India Annamacharya University Department of ECE Rajampet516126 India Sreenivasa Institute of Technology and Management Studies Department of Computer Science and Engineering Chittoor517127 India Anurag University School of Engineering Department of Computer Science and Engineering Hyderabad500088 India ICFAI University Tech School Computer Science Department Ranchi835222 India Amity University Amity School of Engineering and Technology Computer Science Department Noida201313 India Bakhtar University Department of Computer Science and Engineering Karta-e-Chahar Kabul1001 Afghanistan Lovely Professional University Division of Research and Development Phagwara144001 India Chitkara University Institute of Engineering and Technology Chitkara University Centre for Research Impact and Outcome Rajpura140401 India

Visible Light Communication (VLC) has emerged as a promising technology for vehicular communication due to its high data rates, low latency, and immunity to electromagnetic interference. However, optimizing spectral efficiency and channel estimation in dynamic vehicular environments remains a key challenge. This paper proposes an OFDM-QAM-based VLC system to enhance spectral efficiency and improve channel estimation accuracy in vehicular networks. To achieve this, we leverage Simulation of Urban Mobility (SUMO) data, which provides detailed insights into vehicular movement and traffic patterns. By integrating SUMO with VLC system simulations, we optimize modulation schemes, multiple access techniques, and spatial reuse strategies to adapt to varying traffic conditions. Additionally, a mathematical framework is developed to evaluate the impact of vehicular mobility on spectral efficiency and bit error rate (BER) performance across different QAM modulation schemes. Simulation results demonstrate that the proposed VLC system significantly improves spectral efficiency, achieving values up to 1.4 bits/s/Hz, while maintaining low BER across varying Signal to Noise Ratio (SNR) levels. These findings confirm the effectiveness of the proposed methodology in enhancing Visible Light Communication (VLC) performance for vehicular applications, making it a viable solution for next-generation intelligent transportation systems. © 2013 IEEE.

关键词： Spectrum efficiency

来源：评论

学校读者我要写书评

暂无评论

Robust WiFi Sensing-Based Human Pose Estimation Using Denoising Autoencoder and CNN With Dynamic Subcarrier Attention

引用

IEEE Internet of Things Journal 2025年第11期12卷 17066-17079页

作者： Nguyen, Xuan Hoang Nguyen, Van-Dinh Luu, Quang-Trung Gian, Toan Dinh Shin, Oh-Soon Center for Environmental Intelligence College of Engineering and Computer Science Hanoi100000 Viet Nam Hanoi University of Science and Technology School of Electrical and Electronic Engineering Hanoi100000 Viet Nam Soongsil University School of Electronic Engineering Seoul Korea Republic of

WiFi sensing-based human pose estimation (HPE) has gained significant attention in the academic community due to its advantages over vision-and sensor-based methods, including non-intrusiveness, convenience, and enhanced privacy protection. However, most existing WiFi-based pose estimators suffer from poor performance and lack robustness in the presence of random noise. To address these challenges, this paper presents a novel HPE architecture comprising two key modules: A Denoiser and an Estimator. The Denoiser is based on an autoencoder structure, while the Estimator is based on a new convolutional neural network (CNN) called SDy-CNN, which is designed to dynamically focus on high-information subcarriers of orthogonal frequency division multiplexing signals. Additionally, Bayesian optimization is employed to fine-tune the architecture's parameters for optimal performance flexibly. Experiments conducted on a comprehensive dataset, MM-Fi, demonstrate that the proposed architecture significantly outperforms existing state-of-the-art estimators, achieving up to an 8.38% improvement in HPE accuracy in clean data scenarios and up to a 14% improvement in noisy data scenarios. It has also been proven to gain computational efficiency when being much faster than other methods. © 2025 IEEE.

关键词： Wi-Fi

来源：评论

学校读者我要写书评

暂无评论

Scalable Motion Estimation and Temporal Context Reinforcement for Video Compression Using RGB Sensors

引用

IEEE Sensors Journal 2025年第10期25卷 18323-18333页

作者： Huang, Qian Lu, Hao Liu, Wenting Wang, Yiming Hohai University College of Computer Science and Software Engineering Nanjing211100 China Jiangsu Engineering Research Center of Digital Twinning Technology for Key Equipment in Petrochemical Process Changzhou213164 China

We propose a novel deep learned video compression technique, named scalable motion estimation (SME), which is designed for video data generated by sensor systems in smart devices. These devices face unique challenges due to limited bandwidth, constrained computational resources, and the complexity of real-world scenarios. Existing learned video compression methods primarily rely on optical flow for motion estimation;however, optical flow often suffers from inaccuracies and displacement errors, resulting in imprecise motion vectors, and the increasing complexity of models further limits their deployment in resource-constrained environments. To address these issues, the SME framework introduces a hierarchical motion vector enhancement strategy that progressively optimizes motion information in video captured by sensors. In addition, we propose the temporal context reinforcement (TCR) module, which partitions high-dimensional contextual information into three components. Compared to direct context encoding methods, the TCR module optimizes bit allocation, achieving a more balanced distribution of high- and low-frequency information. Extensive experiments on benchmark datasets demonstrate that the SME framework substantially enhances the efficiency and performance of video compression systems on RGB sensor-captured videos, making it particularly suitable for video applications. © 2001-2012 IEEE.

关键词： Optical flows

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：