检索结果-内蒙古大学图书馆

2023 Asia-Pacific Conference on Image Processing, Electronics and computers, IPEC 2023

作者： Sun, Chen Tian, Xiaolin Macau University of Science and Technology Faculty of Innovation Engineering School of Computer Science and Engineering China

Lunar domes have always been one of the important windows to understand lunar volcanic activities, but traditional geological dome identification methods are costly. This study attempts to establish an automatic identification method for lunar volcanic domes through FSOD (Few-Shot Object Detection). Since our previous research has been trying to automatically identify lunar volcanic hills using traditional recognition algorithms, our team will try Few-Shot Object Detection this time. In this study, the researchers first obtained the dome coordinates from the list of known lunar domes and intercepted the data this research need from the corresponding coordinates on the DEM moon images. Subsequently, this research used the existing data to train six traditional recognition algorithms and FSOD and compared their performance to verify the feasibility of this study. This research use AP50 and AP75 to evaluate the performance of the model. Finally, this research found that the AP50 of FSOD can reach 0.76, and the AP75 can reach 0.43. However, although FSOD does perform better than other traditional identification methods, this research believe its accuracy is still far from our expectations. Admittedly, compared with YOLO v5, the performance of FSOD has not been significantly improved, but this study still verifies that it is feasible to apply FSOD to the identification of lunar volcanic domes. © Published under licence by IOP Publishing Ltd.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

DialTest-EA: An Enhanced Fuzzing Approach With Energy Adjustment for Dialogue Systems via Metamorphic Testing

DialTest-EA: An Enhanced Fuzzing Approach With Energy Adjust...

引用

作者： Chen, Haibo Chen, Jinfu Wu, Yucheng Cai, Saihua Ahmad, Bilal Huang, Rubing Wang, Shengran Zhang, Chi School of Computer Science and Communication Engineering Jiangsu Key Laboratory of Security Technology Industrial Cyberspace Jiangsu University Zhenjiang China School of Computer Science and Engineering Faculty of Innovation Engineering Macau University of Science and Technology China

Deep neural networks (DNNs) possess potent feature learning capability, enabling them to comprehend natural language, which strongly support developing dialogue systems. However, dialogue systems usually perform incorrect behaviours in some corner cases, which may cause misunderstanding or economic loss. To test and debug dialogue systems, a popular fuzzing framework by metamorphic testing with Gini impurity guidance is proposed, namely, DialTest. However, DialTest treats all seeds (the initial test inputs to generate the mutated test inputs) equally during the fuzzing process and does not differentiate seeds, resulting in a certain limitation to its incorrect behaviour detection capability. In this paper, we propose to enhance the DialTest by applying a lightweight energy adjustment strategy called DialTest with Energy Adjustment (DialTest-EA). DialTest-EA employs the ant colony optimization algorithm (ACO) to adjust the mutation energy of each seed adaptively, ensuring that potential seeds have more opportunities to generate subsequent test inputs. To evaluate the effectiveness of the proposed DialTest-EA, we conduct a series of comparisons with the original DialTest and random mutation strategy. The experimental results show that the proposed DialTest-EA outperforms the compared methods both in the intent detection and slot filling tasks. Compared with the original DialTest, the intent detection accuracy of generated test cases by the proposed method is reduced by more than 14%, and the slot filling accuracy is reduced by more than 8%. © 2024 John Wiley & Sons Ltd.

关键词： Ant colony optimization

来源：评论

学校读者我要写书评

暂无评论

Deep Learning Approaches for Battery Capacity and State of Charge Estimation with the NASA B0005 Dataset

引用

computers, Materials & Continua 2025年第6期83卷 4795-4813页

作者： Zeyang Zhou Zachary James Ryan Utkarsh Sharma Tran Tien Anh Shashi Mehrotra Angelo Greco Jason West Mukesh Prasad School of Computer Science Faculty of Engineering and Information TechnologyUniversity of Technology SydneyUltimoNSW 2007Australia School of Data Science and Engineering Indian Institute of Science Education and ResearchBhopal462066India Research Institute of Marine Systems Engineering Seoul National UniversitySeoul08826Republic of Korea Department of Computer Science and Engineering Faculty of Engineering and TechnologySRM Institute of Science and TechnologyDelhi NCR CampusGhaziabad201204India Battery System Modelling Department CLARIOS VARTA Hannover GmbHHanover30419Germany Science and Innovation Group Bureau of MeteorologyBrisbane4000Australia

Accurate capacity and State of Charge(SOC)estimation are crucial for ensuring the safety and longevity of lithium-ion batteries in electric *** study examines ten machine learning architectures,Including Deep Belief Network(DBN),Bidirectional Recurrent Neural Network(BiDirRNN),Gated Recurrent Unit(GRU),and others using the NASA B0005 dataset of 591,458 *** indicate that DBN excels in capacity estimation,achieving orders-of-magnitude lower error values and explaining over 99.97%of the predicted variable’s *** computational efficiency is paramount,the Deep Neural Network(DNN)offers a strong alternative,delivering near-competitive accuracy with significantly reduced prediction *** GRU achieves the best overall performance for SOC estimation,attaining an R^(2) of 0.9999,while the BiDirRNN provides a marginally lower error at a slightly higher computational *** contrast,Convolutional Neural Networks(CNN)and Radial Basis Function Networks(RBFN)exhibit relatively high error rates,making them less viable for real-world battery *** of error distributions reveal that the top-performing models cluster most predictions within tight bounds,limiting the risk of overcharging or deep *** findings highlight the trade-off between accuracy and computational overhead,offering valuable guidance for battery management system(BMS)designers seeking optimal performance under constrained *** work may further explore advanced data augmentation and domain adaptation techniques to enhance these models’robustness in diverse operating conditions.

关键词： Battery capacity estimation state of charge deep learning prediction efficiency energy storage systems

来源：评论

学校读者我要写书评

暂无评论

Transformer and Long short term memory networks for long sequence time sequence forecasting problem 5

Transformer and Long short term memory networks for long seq...

引用

5th International Conference on computer Information science and Artificial Intelligence, CISAI 2022

作者： Fang, Wei School of Computer Science and Engineering Faculty of Innovation Engineering Macau University of Science and Technology 999078 China

ISBN: (纸本)9781510662520

The long sequence time-sequence forecasting problem attracts a lot of organizations. Many prediction application scenes are about long sequence time-sequence forecasting problems. Under such circumstances, many researchers have tried to solve these problems by employing some models that have proved efficient in the Natural Language Processing field, like long short term memory networks and Transformers, etc. And there are a lot of improvements based on the primary recurrent neural network, and Transformer. Recently, a model called informer which is made for the LSTF was proposed. This model claimed that it improves prediction performance on the long sequence time-series forecasting problem. But in the later experiments, more and more researchers found that informers still can not handle all the long sequence time-sequence forecasting problems. This paper is going to look at how datasets effect the performance of different models. The experiment is carried out on the Bitcoin dataset with four features and one output. The result shows that the Informer (transformer-like model) cannot always perform well so that sometimes choosing models with simple architecture may gain better results. © 2023 SPIE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

A Deep Learning Approach for Lunar Impact Crater Detection Based on YOLO v7 and CBAM Attention Mechanism 8

A Deep Learning Approach for Lunar Impact Crater Detection B...

引用

8th International Conference on Intelligent Computing and Signal Processing, ICSP 2023

作者： Zhu, Jionghao Liang, Jiarui Tian, Xiaolin Yan, Pengcheng Macau University of Science and Technology School of Computer Science and Engineering Faculty of Innovation Engineering Macau China

ISBN: (纸本)9798350302455

Impact craters are the most common geomorphic unit on the lunar surface, and there are a large number of impact craters of different sizes and morphologies on the lunar surface. Lunar impact craters are a key basis for lunar geological studies, and their study allows for the exploration of the Moon and other planets. Therefore, this paper builds a convolutional neural network YOLO V7-CBAM based on YOLO V7 and the attention mechanism for identifying lunar impact craters. Based on the full-moon CCD images and DEM images provided by NASA, a comparison test was conducted and the precision of YOLO V7-CBAM based on both images was higher than that of YOLO V7, from 72.31% to 74.65% based on CCD images;and from 70.60% to 71.70% based on DEM images. © 2023 IEEE.

关键词： Object detection

来源：评论

学校读者我要写书评

暂无评论

Dictionary Attention-Weighted Cross-Domain Contrastive Learning for Remote Sensing Image Change Detection

引用

Data Intelligence 2024年第4期6卷 893-908页

作者： Tian Wei Youfa Liu Rui Zhao School of Computer Science Wuhan UniversityWuhan430072China Faculty of Electrical Engineering and Computer Science Ningbo University315211China

Significant progress has been made in remote sensing image change detection due to the rapid development of Deep Learning techniques. Convolutional neural networks(CNNs) have become foundational models in this field. Previous works on remote sensing image change detection has utilized domain adaptation methods, achieving promising predictive performance. However, the transferable knowledge between source and target domain has not been fully exploited. In this paper, we propose a novel cross-domain contrastive learning approach for remote sensing image change detection, which correlates source and target domain using contrastive principles. Specifically, we introduce a transferable cross-domain Dictionary Learning scheme where a shared dictionary between the source and target domains generates sparse representations. Based on these representations, we compute attention weights and propose an attention-weighted contrastive loss to enhance knowledge transfer between source and target domains. Experiments demonstrate the effectiveness of the proposed methods on public remote sensing image change detection datasets.

关键词： Change detection Domain adaptation Contrastive learning Dictionary Learning Attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Sifter: An Efficient Operator Auto-Tuner with Speculative Design Space Exploration for Deep Learning Compiler

引用

IEEE Transactions on computers 2024年 1-12页

作者： Zhao, Qianhe Wang, Rui Liu, Yi Yang, Hailong Luan, Zhongzhi Qian, Depei Hangzhou Innovation Institute Beihang University Hangzhou China School of Computer Science and Engineering Beihang University Beijing China

Deep learning compiler can automatically optimize operators. It provides higher flexibility compared to vendor libraries. However, existing DNN operator tuning methods mostly rely on search-based approaches, which still face challenges such as large design spaces and long tuning times. To address these issues, we propose Sifter, an efficient DNN operator auto-tuner with speculative design space exploration. By training and analyzing decision trees, we extract shared characteristics of high-quality schedules and summarize them as pruning rules. Applying these rules during the optimization allows us to speculatively explore the design space, minimize unnecessary hardware measurements, and shorten the optimization time without compromising the optimization result. We conducted experiments on three different platforms with various operators and models. The results demonstrate that Sifter reduces 52% of redundant schedules and shortens the optimization time by 41% while maintaining operator optimization performance at the state-of-the-art level. IEEE

关键词： Space research

来源：评论

学校读者我要写书评

暂无评论

Machine learning-assisted prediction and optimization of solid oxide electrolysis cell for green hydrogen production

引用

Green Chemical engineering 2025年第2期6卷 154-168页

作者： Qingchun Yang Lei Zhao Jingxuan Xiao Rongdong Wen Fu Zhang Dawei Zhang School of Chemistry and Chemical Engineering Hefei University of TechnologyHefei230009China East China Engineering Science and Technology Co.Ltd. Hefei230011China Faculty of Innovation Engineering Macao University of Science and TechnologyMacao999078China

The solid oxide electrolysis cell(SOEC)holds great promise to efficiently convert renewable energy into ***,traditional modeling methods are limited to a specific or reported SOEC ***,four machine learning models are developed to predict the performance of SOEC processes of various types,operating parameters,and feed *** impact of these features on the SOEC's outputs is explained by the Shapley additive explanations and partial dependency plot *** preferredmodel is integratedwith a genetic algorithmto determine the optimal values of each input *** show the improved extreme gradient enhanced regression(XGBoost)algorithm is the core of the machine learning model of the process since it has the highest R^(2)(>0.95)in the three *** electrolytic cell descriptors have a greater impact on the system performance,contributing up to 54.5%.The effective area,voltage,and temperature are the three most influential factors in the SOEC system,contributing 21.6%,16.6%,and 13.0%to its *** temperature,high pressure,and low effective area are the most favorable conditions for H_(2)production *** conducting multi-objective optimization,the optimal current intensity and hydrogen production rate were determined to be 1.61 A/cm^(2)and 1.174 L/(h⋅cm^(2)).

关键词： Solid oxide electrolysis cell Machine learning H_(2)production rate Feature importance analysis System optimization

来源：评论

学校读者我要写书评

暂无评论

Optimizing Stand-Alone PV Systems:A Metaheuristic-Enhanced Fuzzy Approach for Adaptive MPPT

引用

China Communications 2025年第1期22卷 61-74页

作者： Tina Samavat Mostafa Nazari Lin Fuhong Lei Yang Department of Computer Science and Engineering University of NevadaRenoNVUSA Faculty of Mechanical Engineering Shahrood University of TechnologyShahroodIran School of Computer and Communication Engineering University of Science and Technology BeijingBeijing 100083China

This paper introduces a simple yet effective approach for developing fuzzy logic controllers(FLCs)to identify the maximum power point(MPP)and optimize the photovoltaic(PV)system to extract the maximum power in different environmental *** propose a robust FLC with low computational complexity by reducing the number of membership functions and *** optimize the performance of the FLC,metaheuristic algorithms are employed to determine the parameters of the *** evaluate the proposed FLC in various panel configurations under different environmental *** results indicate that the proposed FLC can easily adapt to various panel configurations and perform better than other benchmarks in terms of enhanced stability,responsiveness,and power transfer under various scenarios.

关键词： genetic algorithm imperialist competitive algorithm invasive weed algorithm maximum power point tracking

来源：评论

学校读者我要写书评

暂无评论

IRS-enabled NOMA communication systems:A network architecture primer with future trends and challenges

引用

Digital Communications and Networks 2024年第5期10卷 1503-1528页

作者： Haleema Sadia Ahmad Kamal Hassan Ziaul Haq Abbas Ghulam Abbas Muhammad Waqas Zhu Han Faculty of Electrical Engineering and Telecommunications and Networking(TeleCoN)Research Center Ghulam Ishaq Khan Institute of Engineering and Technology(GIKI)Topi 23640Pakistan Faculty of Computer Science and Engineering GIKITopi 23640Pakistan Department of Computer Engineering College of Information TechnologyUniversity of Bahrain32028Bahrain School of Engineering Edith Cowan UniversityPerth WA6027Australia Department of Electrical and Computer Engineering University of HoustonHoustonTX 77004USA Department of Computer Science and Engineering Kyung Hee UniversitySeoul 446-701South Korea

Non-Orthogonal Multiple Access(NOMA)has already proven to be an effective multiple access scheme for5th Generation(5G)wireless *** provides improved performance in terms of system throughput,spectral efficiency,fairness,and energy efficiency(EE).However,in conventional NOMA networks,performance degradation still exists because of the stochastic behavior of wireless *** combat this challenge,the concept of Intelligent Reflecting Surface(IRS)has risen to prominence as a low-cost intelligent solution for Beyond 5G(B5G)*** this paper,a modeling primer based on the integration of these two cutting-edge technologies,i.e.,IRS and NOMA,for B5G wireless networks is *** in-depth comparative analysis of IRS-assisted Power Domain(PD)-NOMA networks is provided through 3-fold ***,a primer is presented on the system architecture of IRS-enabled multiple-configuration PD-NOMA systems,and parallels are drawn with conventional network configurations,i.e.,conventional NOMA,Orthogonal Multiple Access(OMA),and IRS-assisted OMA *** by this,a comparative analysis of these network configurations is showcased in terms of significant performance metrics,namely,individual users'achievable rate,sum rate,ergodic rate,EE,and outage ***,for multi-antenna IRS-enabled NOMA networks,we exploit the active Beamforming(BF)technique by employing a greedy algorithm using a state-of-the-art branch-reduceand-bound(BRB)*** optimality of the BRB algorithm is presented by comparing it with benchmark BF techniques,i.e.,minimum-mean-square-error,zero-forcing-BF,and ***,we present an outlook on future envisioned NOMA networks,aided by IRSs,i.e.,with a variety of potential applications for 6G wireless *** work presents a generic performance assessment toolkit for wireless networks,focusing on IRS-assisted NOMA *** comparative analysis provides a solid foundation for the dev

关键词： Intelligent reflecting surface Non-orthogonal multiple access 6G Beamforming Sum rate Energy efficiency

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：