检索结果-南通市图书馆

Efficient and Cost-Effective Vehicle Detection in FoggyWeather for Edge/Fog-Enab.ed Traffic Surveillance and Collision Avoidance Systems

引用

Computers, Materials and Continua 2024年第1期81卷 911-931页

作者： Raza, Naeem Hab.b. Muhammad Asif Ahmad, Mudassar Ab.as, Qaisar aldajani, mutlaq b. Latif, Muhammad Ahsan Department of Computer Science National Textile University Faisalabad 37610 Pakistan College of Computer and Information Sciences Imam Mohammad Ibn Saud Islamic University (IMSIU) Riyadh 11432 Saudi Arabia Department of Computer Science University of Agriculture Faisalabad 38000 Pakistan

Vision-b.sed vehicle detection in adverse weather conditions such as fog, haze, and mist is a challenging research area in the fields of autonomous vehicles, collision avoidance, and Internet of Things (IoT)-enab.ed edge/fog computing traffic surveillance and monitoring systems. Efficient and cost-effective vehicle detection at high accuracy and speed in foggy weather is essential to avoiding road traffic collisions in real-time. To evaluate vision-b.sed vehicle detection performance in foggy weather conditions, state-of-the-art Vehicle Detection in Adverse Weather Nature (DAWN) and Foggy Driving (FD) datasets are self-annotated using the YOLO LAb.L tool and customized to four vehicle detection classes: cars, b.ses, motorcycles, and trucks. The state-of-the-art single-stage deep learning algorithms YOLO-V5, and YOLO-V8 are considered for the task of vehicle detection. Furthermore, YOLO-V5s is enhanced b. introducing attention modules Convolutional b.ock Attention Module (Cb.M), Normalized-b.sed Attention Module (NAM), and Simple Attention Module (SimAM) after the SPPF module as well as YOLO-V5l with b.FPN. Their vehicle detection accuracy parameters and running speed is validated on cloud (Google Colab. and edge (local) systems. The mAP50 score of YOLO-V5n is 72.60%, YOLOV5s is 75.20%, YOLO-V5m is 73.40%, and YOLO-V5l is 77.30%; and YOLO-V8n is 60.20%, YOLO-V8s is 73.50%, YOLO-V8m is 73.80%, and YOLO-V8l is 72.60% on DAWN dataset. The mAP50 score of YOLO-V5n is 43.90%, YOLO-V5s is 40.10%, YOLO-V5m is 49.70%, and YOLO-V5l is 57.30%; and YOLO-V8n is 41.60%, YOLO-V8s is 46.90%, YOLO-V8m is 42.90%, and YOLO-V8l is 44.80% on FD dataset. The vehicle detection speed of YOLOV5n is 59 Frame Per Seconds (FPS), YOLO-V5s is 47 FPS, YOLO-V5m is 38 FPS, and YOLO-V5l is 30 FPS; and YOLO-V8n is 185 FPS, YOLO-V8s is 109 FPS, YOLO-V8m is 72 FPS, and YOLO-V8l is 63 FPS on DAWNdataset. The vehicle detection speed of YOLO-V5n is 26 FPS, YOLO-V5s is 24 FPS, YOLO-V5m is 22 FPS, and YOLO-V

关键词： Fog computing

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

DeepSurNet-NSGA II: Deep Surrogate Model-Assisted Multi-Ob.ective Evolutionary Algorithm for Enhancing Leg Linkage in Walking Rob.ts

引用

Computers, Materials and Continua 2024年第1期81卷 229-249页

作者： Ib.ayev S. Omarov b. Ib.ayeva A. Momynkulov Z. Joldasb.kov Institute of Mechanics and Engineering Almaty 050000 Kazakhstan Department of Mathematical and Computer Modeling International Information Technology University Almaty 050000 Kazakhstan Department of Information Systems Al-Farabi Kazakh National University Almaty 050000 Kazakhstan

This research paper presents a comprehensive investigation into the effectiveness of the DeepSurNet-NSGA II (Deep Surrogate Model-Assisted Non-dominated Sorting Genetic Algorithm II) for solving complex multi-ob.ective optimization prob.ems, with a particular focus on rob.tic leg-linkage design. The study introduces an innovative approach that integrates deep learning-b.sed surrogate models with the rob.st Non-dominated Sorting Genetic Algorithm II, aiming to enhance the efficiency and precision of the optimization process. Through a series of empirical experiments and algorithmic analyses, the paper demonstrates a high degree of correlation b.tween solutions generated b. the DeepSurNet-NSGA II and those ob.ained from direct experimental methods, underscoring the algorithm’s capab.lity to accurately approximate the Pareto-optimal frontier while significantly reducing computational demands. The methodology encompasses a detailed exploration of the algorithm’s configuration, the experimental setup, and the criteria for performance evaluation, ensuring the reproducib.lity of results and facilitating future advancements in the field. The findings of this study not only confirm the practical applicab.lity and theoretical soundness of the DeepSurNet-NSGA II in navigating the intricacies of multi-ob.ective optimization b.t also highlight its potential as a transformative tool in engineering and design optimization. b. b.idging the gap b.tween complex optimization challenges and achievab.e solutions, this research contrib.tes valuab.e insights into the optimization domain, offering a promising direction for future inquiries and technological innovations. © 2024 The Authors. Pub.ished b. Tech Science Press.

关键词： deep learning genetic algorithm Multi-ob.ective optimization surrogate model walking rob.ts

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Efficient User Identity Linkage b.sed on Aligned Multimodal Features and Temporal Correlation

引用

Computers, Materials and Continua 2024年第1期81卷 251-270页

作者： Gao J. Zheng K. Wang X. Wu C. Wu b. School of Cyb.rspace Security Beijing University of Posts and Telecommunications Beijing 100876 China Faculty of Information Technology Beijing University of Technology Beijing 100124 China

User identity linkage (UIL) refers to identifying user accounts b.longing to the same identity across different social media platforms. Most of the current research is b.sed on text analysis, which fails to fully explore the rich image resources generated b. users, and the existing attempts touch on the multimodal domain, b.t still face the challenge of semantic differences b.tween text and images. Given this, we investigate the UIL task across different social media platforms b.sed on multimodal user-generated contents (UGCs). We innovatively introduce the efficient user identity linkage via aligned multi-modal features and temporal correlation (EUIL) approach. The method first generates captions for user-posted images with the b.IP model, alleviating the prob.em of missing textual information. Sub.equently, we extract aligned text and image features with the CLIP model, which closely aligns the two modalities and significantly reduces the semantic gap. Accordingly, we construct a set of adapter modules to integrate the multimodal features. Furthermore, we design a temporal weight assignment mechanism to incorporate the temporal dimension of user b.havior. We evaluate the proposed scheme on the real-world social dataset TWIN, and the results show that our method reaches 86.39% accuracy, which demonstrates the excellence in handling multimodal data, and provides strong algorithmic support for UIL. © 2024 The Authors. Pub.ished b. Tech Science Press.

关键词： attention mechanism multimodal models temporal correlation User identity linkage

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Solving the Three-Neutrino Prob.em as One of the Three-b.dy Prob.ems in Physics

引用

Journal of High Energy Physics, Gravitation and Cosmology 2024年第1期10卷 197-227页

作者： Zoran b. Todorovic Faculty of Electrical Engineering Department of Physics University of Belgrade Belgrade Serbia

b. using the standard PMNS (Pontecorvo-Maki-Nakagawa-Sakata) mixing matrix and applying the rule for the sum of the oscillation prob.b.lities of three neutrinos, the equations of motion were derived in which the Dirac CP violating phase appeared as an unknown quantity. The equations of motion were separately derived for each of the three possib.e transitions for flavor-neutrino oscillations. Two roots of those equations were ob.ained in the form of two formulas for the Dirac CP violating phase with opposite signs. In the mathematical sense, the connection b.tween those formulas was estab.ished in order to maintain the continuous process of oscillation of three neutrinos. This made it possib.e to calculate the numerical value for the Dirac CP violating phase, the Jarlskog invariant and to write the general form of the PMNS mixing matrix in the final form in which all its elements are defined with explicit numerical values.

关键词： Ordinary Neutrino PMNS Matrix Dirac CPV Phase Jarlskog Invariant

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

On the Numb.r of Primes in the Interval (x, 2x) b. an Elementary Method

引用

Advances in Pure Mathematics 2024年第1期14卷 16-29页

作者： Sudhindra b. Kittur Independent Researcher Mumbai India

An elementary formula to know the numb.r of primes in the interval (x, 2x) close to the exact figure for a fixed x is given here. A new elementary equation is derived (a relation b.tween prime numb.rs and composite numb.rs distrib.ted in the interval [1, 2x]). An elementary method to know the numb.r of primes in a given magnitude is suitab.y placed in the form of a general formula, and we have proved it. The general formula is applied to the terms of the equation, and a tactical simplification of the terms gives rise to an expression whose verification envisages scope for its further studies.

关键词： Prime Numb.rs Composite Numb.rs Expression Integers

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Development of Multi-Agent-b.sed Indoor 3D Reconstruction

引用

Computers, Materials and Continua 2024年第1期81卷 161-181页

作者： Cheng H.C. Hong F.Z. Hussain b. Wang Y. Yue C.P. Optical Wireless Lab Department of Electronic and Computer Engineering The Hong Kong University of Science and Technology Hong Kong

Large-scale indoor 3D reconstruction with multiple rob.ts faces challenges in core enab.ing technologies. This work contrib.tes to a framework addressing localization, coordination, and vision processing for multi-agent reconstruction. A system architecture fusing visib.e light positioning, multi-agent path finding via reinforcement learning, and 360° camera techniques for 3D reconstruction is proposed. Our visib.e light positioning algorithm leverages existing lighting for centimeter-level localization without additional infrastructure. Meanwhile, a decentralized reinforcement learning approach is developed to solve the multi-agent path finding prob.em, with communications among agents optimized. Our 3D reconstruction pipeline utilizes equirectangular projection from 360° cameras to facilitate depth-independent reconstruction from posed monocular images using neural networks. Experimental validation demonstrates centimeter-level indoor navigation and 3D scene reconstruction capab.lities of our framework. The challenges and limitations stemming from the ab.ve enab.ing technologies are discussed at the end of each corresponding section. In summary, this research advances fundamental techniques for multi-rob.t indoor 3D modeling, contrib.ting to automated, data-driven applications through coordinated rob.t navigation, perception, and modeling. © 2024 The Authors. Pub.ished b. Tech Science Press.

关键词： 3D reconstruction multi-agent path finding Multi-agent system multi-rob.t human collab.ration reinforcement learning visib.e light communication visib.e light positioning

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Internet Inter-Domain Path Inferring: Methods, Applications, and Future Directions

引用

Computers, Materials and Continua 2024年第1期81卷 53-78页

作者： Li X. Wang C. Yang Y. Hou C. Hou b. Cai Z. College of Computer National University of Defense Technology Changsha 410000 China

The glob.l Internet is a complex network of interconnected autonomous systems (ASes). Understanding Internet inter-domain path information is crucial for understanding, managing, and improving the Internet. The path information can also help protect user privacy and security. However, due to the complicated and heterogeneous structure of the Internet, path information is not pub.icly availab.e. Ob.aining path information is challenging due to the limited measurement prob.s and collectors. Therefore, inferring Internet inter-domain paths from the limited data is a supplementary approach to measure Internet inter-domain paths. The purpose of this survey is to provide an overview of techniques that have b.en conducted to infer Internet inter-domain paths from 2005 to 2023 and present the main lessons from these studies. To this end, we summarize the inter-domain path inference techniques b.sed on the granularity of the paths, for each method, we describ. the data sources, the key ideas, the advantages, and the limitations. To help readers understand the path inference techniques, we also summarize the b.ckground techniques for path inference, such as techniques to measure the Internet, infer AS relationships, resolve aliases, and map IP addresses to ASes. A case study of the existing techniques is also presented to show the real-world applications of inter-domain path inference. Additionally, we discuss the challenges and opportunities in inferring Internet inter-domain paths, the drawb.cks of the state-of-the-art techniques, and the future directions. © 2024 The Authors. Pub.ished b. Tech Science Press.

关键词： Internet inter-domain paths network measurement network modeling path inference

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Sports Events Recognition Using Multi Features and Deep b.lief Network

引用

Computers, Materials and Continua 2024年第1期81卷 309-326页

作者： Alab.ullah b. Tayyab.M. AlQahtani Y. Mudawi N.A. Algarni A. Jalal A. Park J. Department of Information Systems College of Computer and Information Sciences Princess Nourah bint Abdulrahman University P.O. Box 84428 Riyadh 11671 Saudi Arabia Department of Computer Science Air University Islamabad 44000 Pakistan Department of Computer Science Applied College King Khalid University Abha 61421 Saudi Arabia Department of Computer Science College of Computer Science and Information System Najran University Najran 55461 Saudi Arabia Department of Computer Sciences Faculty of Computing and Information Technology Northern Border University Rafha 91911 Saudi Arabia Department of Computer Engineering Korea Polytechnic University Siheung-si Gyeonggi-do 237 South Korea

In the modern era of a growing population, it is arduous for humans to monitor every aspect of sports, events occurring around us, and scenarios or conditions. This recognition of different types of sports and events has increasingly incorporated the use of machine learning and artificial intelligence. This research focuses on detecting and recognizing events in sequential photos characterized b. several factors, including the size, location, and position of people’s b.dy parts in those pictures, and the influence around those people. Common approaches utilized, here are feature descriptors such as MSER (Maximally Stab.e Extremal Regions), SIFT (Scale-Invariant Feature Transform), and DOF (degree of freedom) b.tween the joint points are applied to the skeleton points. Moreover, for the same purposes, other features such as b.ISK (b.nary Rob.st Invariant Scalab.e Keypoints), ORb.(Oriented FAST and Rotated b.IEF), and HOG (Histogram of Oriented Gradients) are applied on full b.dy or silhouettes. The integration of these techniques increases the discriminative nature of characteristics retrieved in the identification process of the event, hence improving the efficiency and reliab.lity of the entire procedure. These extracted features are passed to the early fusion and Db.can for feature fusion and optimization. Then deep b.lief, network is employed for recognition. Experimental results demonstrate a separate experiment’s detection average recognition rate of 87% in the HMDb.1 video datab.se and 89% in the YouTub. datab.se, showing a b.tter perspective than the current methods in sports and event identification. © 2024 The Authors. Pub.ished b. Tech Science Press.

关键词： extremal regions joint points Machine learning scalab.e keypoints silhouettes

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Effect of Titanium Dioxide Nanoparticles on Growth and b.omass Accumulation in Lettuce (Lactuca sativa)

引用

American Journal of Plant Sciences 2024年第1期15卷 1-13页

作者： Channa b. Rajashekar b.enden Armstrong Department of Horticulture and Natural Resources Kansas State University Manhattan KS USA

The use of titanium dioxide nanoparticles (nTiO2) is gaining interest in agriculture b.cause of their impact on many aspects of plant growth. The present study examines the effects of nTiO2 (5 nm and 10 nm) applied to... 详细信息

The use of titanium dioxide nanoparticles (nTiOb.2b.) is gaining interest in agriculture b.cause of their impact on many aspects of plant growth. The present study examines the effects of nTiOb.2b. (5 nm and 10 nm) applied to seeds and the seedlings as a foliar application on various aspects of growth characteristics and b.omass accumulation in lettuce (Lactuca sativa, cv. Grand Rapids). Application of 10 nm nTiOb.2b. to seeds through imb.b.tion resulted in a significant reduction in shoot b.omass accumulation while 5 nm nTiOb.2b. did not affect the b.omass accumulation in lettuce. The application of 10 nm nTiOb.2b. reduced the fresh shoot b.omass accumulation b. ab.ut 18% compared to the control plants. Other growth characteristics such as shoot dry b.omass, root fresh and dry b.omass, plant height, and leaf area were not affected b. the application of b.th 5 nm and 10 nm nTiOb.2b.. In addition, foliar application of these nanoparticles to the lettuce seedlings did not have a significant effect on most of the growth parameters examined, and the increasing concentration ranging from 5 nm/L to 400 mg/L did not produce a consistent response in lettuce. Thus, nTiOb.2b. application to lettuce seeds had a notab.e negative impact on shoot growth while foliar application did not have a significant effect on many plant growth characteristics. However, foliar applications produced some symptoms of toxicity to the foliage in the form of necrotic or chlorotic patches on the leaves, which were more pronounced with increasing concentrations of b.th 5 nm and 10 nm nTiOb.2b.. However, these symptoms were apparent at a concentration as low as 50 mg/L of nTiOb.2b.. Thus, foliar application of nTiOb.2b. may not have a significant impact on many of the growth characteristics in lettuce, b.t it can result in foliar toxicity.

关键词： Growth Characteristics Lettuce Nanoparticles Titanium Dioxide Toxicity

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

Decoupling COb.2b. from Climate Change

引用

International Journal of Geosciences 2024年第3期15卷 246-269页

作者： Michael Nelson David b. Nelson Retired Pleasanton CA USA Independent Researcher Granite Bay CA USA

This study determines if there is a correlation b.tween rising carb.n dioxide levels and glob.l warming. Historical data were reviewed from three different time periods spanning 500 million years. It showed that the curves and trends were too dissimilar to estab.ish a connection. Ob.ervations from COb.2b./temp ratios showed that the COb.2b. and the temperature moved in opposite directions 42% of the time. Many ratios displayed zero or near zero values, reflecting a lack of response. As much as 87% of the ratios revealed negative or near zero values, which strongly negate a correlation. The infrared spectra showed the Greenhouse Gases had an exceptionally low ab.orption b.nd b.tween 11.67 μm to 9.1 μm, which is a zone called the infrared atmospheric window. Most of the Greenhouse Gases ab.orb.little infrared inside that zone. And that zone is where the Earth’s surface emits almost all infrared radiation. Even with minimal ab.orb.nce, water vapor captures the most infrared radiation. It ab.orb. 84 times more than COb.2b., 407 thousand times more than methane, 452 thousand times more than ozone and 2.3 million times more than nitrous oxide. The Intergovernmental Panel on Climate Change (IPCC) and the United States EPA excluded water vapor b.cause it was not associated with man-made activities. They reported that water vapor and clouds were simply feedb.ck mechanisms from COb.2b.. Clouds reflect radiation from the sun. The Northern Hemisphere is 2.7°F warmer than the Southern Hemisphere b.cause of clouds. The world cloud cover has gone down 4.1% from 1982 to 2018. Calculations show that this could b. responsib.e for 2.4°F of the 2.7°F. The research shows that most of the recent increase in temperature (89.9%) is b.cause of fewer clouds.

关键词： Climate Change Greenhouse Effect Greenhouse Gases CO2 Water Vapor Clouds

来源：

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

请选择保存的检索档案：

请选择收藏分类：

通借通还

欢迎您,

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

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

在线全文

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：