检索结果-内蒙古大学图书馆

Improving YOLOv5s Algorithm for Detecting Flame and Smoke

IEEE ACCESS 2024年 12卷 126568-126576页

作者： Deng, Li Zhou, Jin Liu, Quanyi Civil Aviat Flight Univ China Coll Civil Aviat Safety Engn Guanghan 618307 Sichuan Peoples R China Civil Aircraft Fire Sci & Safety Engn Key Lab Sich Guanghan 618307 Sichuan Peoples R China Sichuan Key Technol Engn Res Ctr All Elect Navigab Guanghan 618307 Sichuan Peoples R China

Object detection methods can be used to detect flames and smoke from images or videos for the identification and exploration of fire. In this paper, an improved YOLOv5s algorithm, called GAM-ASFF-YOLOv5s is proposed, which introduces an attention mechanism and feature fusion layer. A global attention mechanism (GAM) is introduced into the backbone network of YOLOv5s to focus on the detection information that is conducive to flames and smoke, while suppressing unimportant information. A head network with adaptive spatial feature fusion (ASFF) was designed to extract more complete image features of flame and smoke. In addition, the original object bounding box regression loss function the complete IoU (CIoU) of YOLOv5s was replaced by repulsion loss to enhance the generalization ability of the model and further improve the detection performance of the flame and smoke. The experimental results show that the precision of the GAM-ASFF-YOLOv5s+REP algorithm was 5.7% higher than that of the original YOLOv5s algorithm on the VOC2007 dataset and it also performed well on the flame and smoke dataset, that is, the precision, recall, and mean average precision (mAP) were improved by 0.5 %, 2.1 % and 2.9 % respectively, compared to the original YOLOv5s algorithm.

关键词： YOLO Feature extraction Fires Convolutional neural networks Accuracy detection algorithms Smoke detectors Adaptive spatial feature fusion flame and smoke detection global attention mechanism repulsion loss yolov5s

来源：评论

学校读者我要写书评

暂无评论

Network Security Perception System Integrating Improved CNN Algorithm and Improved GRU Algorithm

引用

IEEE ACCESS 2024年 12卷 163805-163818页

作者： Huang, Yu Zhao, Ziyi Shi, Canbin China Southern Power Grid Co Ltd EHV Transmiss Co Guangzhou 510663 Peoples R China Jinan Univ Coll Phys & Optoelect Engn Guangzhou 510632 Peoples R China Beijing Venustech Cybervis Co Ltd Beijing 100193 Peoples R China

To enhance the accuracy and real-time response of network security-aware systems, the convolutional neural network is optimized by introducing exponentially weighted dempster-shafer evidence theory. Additionally, the gated recurrent unit is improved using an adaptive boosting algorithm. A novel network security-aware model integrating both methods is constructed. Experimental results show that the model achieves a detection accuracy of up to 94%, with the lowest values for miss rate, false positive rate, and false alarm rate being 7.84%, 2.47%, and 3.11%, respectively. This indicates that the proposed model provides an efficient and accurate solution for network security defense.

关键词： Network security Convolutional neural networks Accuracy Mathematical models Hidden Markov models Adaptation models Real-time systems Heuristic algorithms detection algorithms Convolutional neural network gated recurrent unit network security perception detection

来源：评论

学校读者我要写书评

暂无评论

An Investigation Into the Advancements of Edge-AI Capabilities for Structural Health Monitoring

引用

IEEE ACCESS 2024年 12卷 25325-25345页

作者： Mishra, Anoop Gangisetti, Gopinath Khazanchi, Deepak Univ Nebraska Coll Informat Sci & Technol Omaha NE 68182 USA

The aim of this study is to investigate the capabilities of edge-AI in the field of structural health monitoring, with a particular emphasis on detecting cracks in concrete bridges. Comprehensive literature suggests that edge-AI approaches have not been utilized in the structural health monitoring domain (SHM). This article proposed two novel frameworks: an edge-AI framework for the SHM domain and a cloud-edge adaptive intelligence for crack detection (CEAIC) to utilize edge-AI approaches for real-time scenarios. The framework incorporates a novel edge-AI framework, the crowd intelligence approach, and the CrANET framework to perform weakly supervised crack segmentation. Quantization approaches are used to transform the deep learning model into an optimized model to be compatible with edge devices. The edge-AI experiments for the cracks detection task are conducted using Kneron KL520 and Google Coral development board. A responsive website has been developed to demonstrate the real-world implications of the CEAIC framework. This study has the potential to provide a cost-effective and reliable solution for real-time monitoring and assessment of concrete bridge cracks, thereby improving the safety and longevity of bridges. The outcomes of this research endeavor will furnish us with invaluable insights regarding the feasibility and potential advantages of incorporating edge AI into the SHM domain.

关键词： Image edge detection Bridges Monitoring Cloud computing Task analysis Computational modeling Real-time systems Edge computing Artificial intelligence Neural networks Quantization (signal) Structural engineering detection algorithms Surface cracks Cracks detection edge-AI neural networks quantization structural health monitoring

来源：评论

学校读者我要写书评

暂无评论

Detecting and Correcting Gain Jumps in TES Microcalorimeters

引用

IEEE TRANSACTIONS ON APPLIED SUPERCONDUCTIVITY 2025年第5期35卷

作者： Baker, Thomas A. Becker, Daniel T. Fowler, Joseph W. Keller, Mark W. Swetz, Daniel S. Ullom, Joel N. Univ Colorado Boulder Dept Phys Boulder CO 80309 USA Natl Inst Stand & Technol Boulder CO 80305 USA

Arrays of microcalorimeters based on transition-edge sensors (TESs) are being actively deployed to laboratories all over the world. A TES microcalorimeter array produces very large quantities of data and users of these devices have varying levels of experience, so it is important to provide robust software for data acquisition and analysis that can function with minimal user supervision. This software should be capable of addressing common phenomena that can adversely affect spectrum quality. Gain jumping is one such phenomenon that is characterized by abrupt changes in the gain of a device. Left unaddressed, gain jumps can degrade spectra by introducing false peaks. We are not aware of any previously published methods for resetting gain jumps during data acquisition or existing algorithms for correcting data that is degraded by gain jumps. We have developed automated methods for detecting and correcting gain jumps in gamma-ray TES microcalorimeters. We present a procedure for resetting gain jumps during a live data acquisition that involves briefly driving the TES into its normal state using the bias current. We also describe an algorithm for locating gain jumps and identifying unique gain states within existing microcalorimeter data. Finally, we provide a possible approach for correcting gain jumps after they have been identified.

关键词： Detectors Standards Superconductivity Arrays Kernel Gamma-rays Data acquisition Shape Noise detection algorithms Data analysis gamma rays detectors spectroscopy

来源：评论

学校读者我要写书评

暂无评论

Optimizing Solid Waste Management: A Holistic Approach by Informed Carbon Emission Reduction

引用

IEEE ACCESS 2024年 12卷 121659-121674页

作者： Hegde, Saumya Sumith, N. Pinto, Twensica Shukla, Shivam Patidar, Vijay Natl Inst Technol Karnataka Dept Comp Sci & Engn Surathkal 575025 India Manipal Acad Higher Educ Manipal Inst Technol Dept Informat & Commun Technol Manipal 576104 Karnataka India

Reducing carbon monoxide (CO) emissions is imperative for safeguarding human health and environment. CO adversely affects respiratory health, contributing to respiratory problems and, in severe cases, fatalities. Its reduction aligns with the broader efforts to combat climate change, as CO is often emitted alongside other greenhouse gases. Environmental consequences include air pollution and its detrimental impact on ecosystems. Compliance with emission standards is essential, and reducing Carbon emissions can lead to social and economic benefits, such as increased productivity and reduced healthcare costs. Moreover, the focus on emission reduction drives technological innovation, fostering the development of cleaner and sustainable technologies. In essence, addressing CO emissions is vital for creating a healthier, more sustainable future. However, in most of the cases, there has been no much importance given in scientific management of solid wastes. This has therefore resulted in large magnitude of carbon emission causing serious implications. This paper presents a novel approach to solid waste management, combining carbon emission assessment with advanced object detection technology. We develop an integrated waste management model that employs machine learning techniques for the identification and categorization of metals, non-metals, and plastics within the solid waste stream. To optimize waste sorting and recycling processes, we implement an efficient object detection system that leverages computer vision algorithms. This system enhances the precision of material identification within solid waste, thereby improving sorting accuracy. Additionally, we establish a database to quantify carbon emissions associated with distinct waste management methods, encompassing incineration, composting, recycling, bioremediation, and landfills is used for this work. The novelty of the work lies in the integration of CO2 emissions data and object detection resulting into a decis

关键词： Carbon monoxide Climate change Waste materials Waste management Greenhouse gases Carbon emissions Sustainable development Environmental monitoring Socioeconomics detection algorithms CNN models carbon estimation waste management solid waste carbon footprint

来源：评论

学校读者我要写书评

暂无评论

AGVS: A New Change detection Dataset for Airport Ground Video Surveillance

引用

IEEE TRANSACTIONS ON INTELLIGENT TRANSPORTATION SYSTEMS 2022年第11期23卷 20588-20600页

作者： Zhang, Xiang Shu, Chang Li, Shuai Wu, Celimuge Liu, Zhi Univ Elect Sci & Technol China Yangtze Delta Reg Inst Quzhou Sch Informat & Commun Engn Chengdu 611731 Peoples R China Shandong Univ Sch Control Sci & Engn Jinan 250100 Peoples R China Univ Electrocommun Dept Comp & Network Engn Tokyo 1828585 Japan

Change detection is the foundation of intelligent video surveillance of the airport ground. However, experiments have shown that change detection algorithms with good performance on traditional datasets (e.g., CDnet2014) perform poorly in airport ground surveillance. The reason is that traditional datasets focus on the diversity of scenarios, while the practical application requires robustness against various changes in a single scene. We posit that the solution to this problem is to establish a unique dataset for airport ground surveillance and develop specific algorithms for this scenario. In this paper, we present an Airport Ground Video Surveillance benchmark (AGVS) for change detection of the airport ground. AGVS includes 25 long videos, amounting to about 100000 frames and accurate ground truth for all frames. Each video contains multiple challenges specific to the airport ground (e.g., haze, camouflage, strip shape, shadow and illumination change, simultaneous multi-scale objects) and various appearance changes of the aircraft). Change detection ground truth is generated by manual annotation. The AGVS benchmark can be downloaded from ***. Furthermore, we conduct a simple review of current change detection algorithms, both unsupervised or supervised, and then 21 state-of-the-art algorithms are tested and analyzed on the AGVS benchmark. Finally, we conclude with algorithm design principles of change detection for airport ground surveillance.

关键词： Airports Change detection algorithms detection algorithms Benchmark testing Adaptation models Video surveillance Lighting Benchmark change detection background subtraction airport ground video surveillance

来源：评论

学校读者我要写书评

暂无评论

A Zero-Velocity detection Algorithm Robust to Various Gait Types for Pedestrian Inertial Navigation

引用

IEEE SENSORS JOURNAL 2022年第6期22卷 4916-4931页

作者： Cho, Seong Yun Lee, Jae Hong Park, Chan Gook Kyungil Univ Dept Robot & Mobil Gyongsan 38428 South Korea Seoul Natl Univ Dept Aerosp Engn ASRI Seoul 08826 South Korea

In this article, we develop a pedestrian inertial navigation (PIN) that uses an inertial measurement unit (IMU) mounted on a shoe to locate a pedestrian in an infrastructure-free environment. In the PIN, stand-alone navigation is possible by calculating the navigation solution using the inertial navigation system (INS) mechanism. For accurate navigation, however, it is necessary to compensate for accumulated errors over time. In a foot-mounted PIN, the error can be compensated through zero-velocity (ZV) correction method by detecting the moment the foot touches the ground. One important point here is to detect ZVs accurately. In most of the literature related to this, ZV detection techniques for normal gait have been proposed. In this article, we first propose a powerful ZV detection algorithm for various gait types. This algorithm consists of a new buffering method of the sensor output power, an exact ZV detection method through glitch removal, and a reinforcement method for ZV detection based on auxiliary step detection. Second, we design a PIN/ZV integration filter, and describe the real-time operation plan of all these functions. The performance of the PIN using the proposed algorithm is verified through experiments.

关键词： Navigation Pins Sensors detection algorithms Legged locomotion Foot Inertial navigation Pedestrian inertial navigation PIN ZV integration zero-velocity detection

来源：评论

学校读者我要写书评

暂无评论

Community Deception in Attributed Networks

引用

IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS 2024年第1期11卷 228-237页

作者： Fionda, Valeria Pirro, Giuseppe Univ Calabria DeMaCS I-87036 Arcavacata Di Rende Italy Sapienza Univ Rome DI I-00185 Rome Italy

Community detection algorithms that analyze networks to identify communities of nodes are an essential part of the network analysis toolkit used daily by different analysts (e.g., data scientists and law enforcement). However, there is not enough awareness that members of a community C (either revealed or not) inside a network $G$ could act strategically to evade such tools either for legitimate (e.g., activist groups in authoritarian regimes) or malicious (e.g., terrorists) purpose. Community deception offers this possibility. By identifying a certain number of C 's member connections to be rewired, community deception algorithms can successfully hide a community that wants to stay below the radar of detection techniques. However, the state-of-the-art deception approaches have focused on networks without attributes, although real-world networks (e.g., Facebook) include attributes (e.g., age and sex) that play a central role in detecting more accurate communities. This article faces three novel challenges introduced when designing deception techniques for networks with attributes. The first concerns how to model and encode attributes most flexibly. The second is about framing attribute-aware community deception as an optimization problem. Finally, the challenge of solving the optimization problem by leveraging network topology and attributes also arises. We leverage a simple way to model network attributes as edge weights, a novel optimization function called community diffusion, and Diffuser a greedy algorithm to optimize diffusion, to solve the above challenges. We evaluated Diffuser against several community detection algorithms and compared it with state-of-the-art deception approaches on various real-world networks. From the evaluation, we can draw two main observations. First, adopting attribute-oblivious deception techniques leads to unsatisfactory results. Second, community diffusion as an optimization function specific to attributed networks is preferred to

关键词： Image edge detection Optimization Detectors detection algorithms Social networking (online) Network analyzers Minimization Social network deception

来源：评论

学校读者我要写书评

暂无评论

Backdoor Attacks to Deep Neural Networks: A Survey of the Literature, Challenges, and Future Research Directions

引用

IEEE ACCESS 2024年 12卷 29004-29023页

作者： Mengara, Orson Avila, Anderson Falk, Tiago H. Univ Quebec INRS EMT Montreal PQ H5A 1K6 Canada INRS UQO Joint Res Unit Cybersecur Gatineau PQ Canada

Deep neural network (DNN) classifiers are potent instruments that can be used in various security-sensitive applications. Nonetheless, they are vulnerable to certain attacks that impede or distort their learning process. For example, backdoor attacks involve polluting the DNN learning set with a few samples from one or more source classes, which are then labeled as target classes by an attacker. Even if the DNN is trained on clean samples with no backdoors, this attack will still be successful if a backdoor pattern exists in the training data. Backdoor attacks are difficult to spot and can be used to make the DNN behave maliciously, depending on the target selected by the attacker. In this study, we survey the literature and highlight the latest advances in backdoor attack strategies and defense mechanisms. We finalize the discussion on challenges and open issues, as well as future research opportunities.

关键词： Artificial neural networks Data models Training Surveys Trojan horses Training data Deep learning detection algorithms Computer security Backdoor attack deep learning vulnerability detection trojan attack neural trojan

来源：评论

学校读者我要写书评

暂无评论

Enhancing Relationship Link Prediction With Hierarchical Feature Enhancement

引用

IEEE ACCESS 2024年 12卷 174387-174398页

作者： Xu, Zhouying Wang, Huiyue Coll Econ & Management Nefu Harbin 150040 Peoples R China State Grid Hebei Elect Power Co Ltd Shijiazhuang 050000 Peoples R China

In this study, we endeavor to assess the potential for establishing friendships between users, a challenge that intersects with the domains of friend recommendation and link prediction. Notably, prevailing models in friend recommendation predominantly consider only the attributes of users, whereas link prediction models primarily emphasize network structure. In this paper, we propose a novel methodology by regarding users' attributes as node-level features, extracting them from tweet text using GPT-style models, and considering the structure of the constructed network as network-level features. The network-level features are obtained through a soft-partitioning overlapping community detection algorithm, namely BIGCLAM(Cluster Affiliation Model for Big Networks). We then integrate the node-level and network-level features to mutually enhance their predictive capabilities. Through rigorous experimental validation across two distinct dataset types-a homogeneous network and a heterogeneous network-our results demonstrate that the fusion of node-level and network-level features significantly improves performance.

关键词： Feature extraction Social networking (online) detection algorithms Semantics Neural networks Generative AI Deep learning Computational modeling Attention mechanisms Vectors Link prediction node-level feature network-level feature

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：