检索结果-内蒙古大学图书馆

Improved bounding box regression loss for weapon detection systems using deep learning

International Journal of Information technology (Singapore) 2024年 1-17页

作者： Sumi, Lucy Dey, Shouvik Department of Computer Science and Engineering National Institute of Technology Nagaland Chumukedima 797103 India

Rising number of crime rate using firearms (such as open firing, robbery, suicides, mass shootings, homicides, threatening at gun point, etc.), has underscored the growing importance of timely detection of weapons. Bounding box regression, an essential element in object detection, plays a crucial role in accurately identifying and localizing these firearms. This paper introduces a Manhattan-Complete IoU (MCIoU) loss for bounding boxes, which demonstrates significantly faster convergence during training compared to other IoU losses. By incorporating MCIoU into one of the advanced object detection framework, (You Only Look Once) YOLOv7, the proposed work demonstrates consistent improvement on their performance across popular weapon detection benchmarks datasets such as Granda, Synthetic, Cam157, Internet Movie Firearms Database (IMFDB), Gun Movie and Monash. The most encouraging outcomes were obtained on a Gun movie dataset with a precision and recall value of 98.2% and 96.3% respectively, which is an appreciable improvement compared to the baseline YOLOv7 model. Experiments show that the best precision achieved was at 98.2% and mAP@.50:.95 of 42.5% over other existing IoUs. © Bharati Vidyapeeth's institute of computer Applications and Management 2024.

关键词： CCTV surveillance Deep learning Intersection over union (IoU) Loss function Weapon detection system You Only Look Once v7 (YOLOv7)

来源：评论

学校读者我要写书评

暂无评论

Credit Card Fraud Detection Using Machine Learning 2

Credit Card Fraud Detection Using Machine Learning

引用

2nd International Conference on Advancements in Smart, Secure and Intelligent Computing, ASSIC 2024

作者： Peneti, Subhashini Rama Krishna, S. Kiran, Ajmeera Tripathy, Hrudaya Kumar Mlr Institute of Technology Department of Computer Science & Information Technology Hyderabad India Gitam University Department of Computer Science & Engineering Hyderabad India Mlr Institute of Technology Department of Computer Science and Engineering Hyderabad India Kalinga Institute of Industrial Technology Deemed to Be University India

ISBN: (纸本)9798350370188

India's banking sector has grown rapidly in modern times. With increased bank functioning, frauds skyrocket. Debit cards, Internet Banking, and ATM cards are among the financial services offered today. These services attracted many new customers worldwide. We present a Machine Learning technique to detect credit card fraud. The global use of credit cards is expanding significantly. Thus, the bank server can track these transactions using Machine Learning. These algorithms detect or anticipate transaction fraud. Each transaction has a few functions in the dataset. We should classify and identify fraudulent transactions from those transactions. The work suggests using Machine Learning calculation for this. © 2024 IEEE.

关键词： Crime

来源：评论

学校读者我要写书评

暂无评论

Towards a blockchain-SDN-based secure architecture for cloud computing in smart industrial IoT

引用

Digital Communications and Networks 2023年第2期9卷 411-421页

作者： Anichur Rahman Md Jahidul Islam Shahab S.Band Ghulam Muhammad Kamrul Hasan Prayag Tiwari Department of Computer Science and Engineering National Institute of Textile Engineering and Research(NITER)Constituent Institute of the University of DhakaSavarDhaka1350Bangladesh Department of Computer Science and Engineering Green University of BangladeshDhakaBangladesh Future Technology Research Center College of FutureNational Yunlin University of Science and TechnologyDouliouTaiwanChina Department of Computer Engineering College of Computer and Information SciencesKing Saud UniversityRiyadhSaudi Arabia Department of Information and Communication Engineering Hankuk University of Foreign StudiesSouth Korea Department of Computer Science Aalto UniversityFinland

Some of the significant new technologies researched in recent studies include BlockChain(BC),Software Defined Networking(SDN),and Smart Industrial Internet of Things(IIoT).All three technologies provide data integrity,confidentiality,and integrity in their respective use cases(especially in industrial fields).Additionally,cloud computing has been in use for several years *** information is exchanged with cloud infrastructure to provide clients with access to distant resources,such as computing and storage activities in the *** are also significant security risks,concerns,and difficulties associated with cloud *** address these challenges,we propose merging BC and SDN into a cloud computing platform for the *** paper introduces“DistB-SDCloud”,an architecture for enhanced cloud security for smart IIoT *** proposed architecture uses a distributed BC method to provide security,secrecy,privacy,and integrity while remaining flexible and *** in the industrial sector benefit from the dispersed or decentralized,and efficient environment of ***,we described an SDN method to improve the durability,stability,and load balancing of cloud *** efficacy of our SDN and BC-based implementation was experimentally tested by using various parameters including throughput,packet analysis,response time,bandwidth,and latency analysis,as well as the monitoring of several attacks on the system itself.

关键词： Smart IIoT Blockchain SDN IoT Security Privacy OpenFlow SDN-Controller Data security Cloud computing Cloud management

来源：评论

学校读者我要写书评

暂无评论

Ontology-Based Semantic Reasoning for Multisource Heterogeneous Industrial Devices Using OPC UA

引用

IEEE Internet of Things Journal 2025年第13期12卷 25020-25032页

作者： Bi, Jing Wu, Rina Yuan, Haitao Wang, Ziqi Zhang, Jia Zhou, MengChu Beijing University of Technology College of Computer Science Beijing100124 China Beihang University School of Automation Science and Electrical Engineering Beijing100191 China Southern Methodist University Lyle School of Engineering Department of Computer Science DallasTX75205 United States New Jersey Institute of Technology Department of Electrical and Computer Engineering NewarkNJ07102 United States

The advent of smart manufacturing in Industry 4.0 signifies the era of connections. As a communication protocol, object linking and embedding for process control unified architecture (OPC UA) can address most semantic heterogeneity issues. However, its semantics are not formally defined at the application layer. To address the information silo problem caused by semantic heterogeneity, an integration framework named querying of ontology mapping-based OPC UA (QOMOU) is proposed. QOMOU extracts information models of OPC UA servers into resource description framework triples and utilizes Web Ontology Language for semantic enrichment and inference. Then, an event class semantic similarity calculation (ECSSC) method is proposed for device-type identification, enabling the classification of semantically heterogeneous OPC UA devices. The effectiveness of ECSSC is validated through queries with the RDF query language (SPARQL) protocol in Apache Jena. Experimental results demonstrate that ECSSC improves the accuracy of device identification by approximately 7% compared to benchmark device identification models. Specifically, compared with graph embedding-based methods, QOMOU’s query performance is approximately 13% higher, and its query efficiency is 5% higher on average compared to both structured query and extensible markup languages. Moreover, by employing a keyword-matching algorithm, the query accuracy of the existing heterogeneous data integration scheme is improved by 4% on average. This enhancement can boost the operational efficiency of Internet of Things systems based on the OPC UA architecture. © 2014 IEEE.

关键词： Structured Query Language

来源：评论

学校读者我要写书评

暂无评论

Automated breast nuclei feature extraction for segmentation in histopathology images using Deep-CNN-based gaussian mixture model and color optimization technique

引用

Multimedia Tools and Applications 2025年 1-27页

作者： Murmu, Anita Kumar, Piyush Computer Science and Engineering National Institute of Technology Patna Ashok Rajpath Bihar Patna800005 India

Hematoxylin and eosin (H&E) staining are the key sources for identifying breast cancer patterns with different colors and shapes of nuclei cells for segmenting histopathology nucleus images. In nucleus cells, the color categorization of hematoxylin is stained blue, whereas cytoplasmic cells of eosin are stained pink. This color dissimilarity in the nucleus cells indicates different structures of the cellular and cancer-specific cells, which need to detect on time is a challenge for accurate diagnosis. To overcome this issue, a Deep Convolution Neural Network (Deep-CNN) based Unet framework with Transfer Learning (TL) is proposed for segmenting histopathology breast cancer images. Specifically, the Deep-CNN framework has been used to extract features automatically from input images, and select the features by using the TL technique for minimizing the target tasks to train it faster. Furthermore, a deep convolutional-based Gaussian Mixture Model (GMM) is used to normalize the shapes and merge the nucleus cell pixels for color optimization. The proposed method has been tested on Kaggle’s 2018 Data Science Blow histopathology breast image dataset. The experimental results of the proposed scheme have been evaluated by performance evaluation matrices, namely accuracy, Jaccard coefficient, Dice Similarity Coefficient (DSC), Intersection over Union (IoU), and F1-score. The proposed approach outperforms compared with different state-of-the-art models on the same dataset. © The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature 2025.

关键词： Convolution

来源：评论

学校读者我要写书评

暂无评论

Hardware Security of Image Processing Cores Against IP Piracy Using PSO-Based HLS-Driven Multi-Stage Encryption Fused with Fingerprint Signature

引用

SN computer Science 2024年第7期5卷 941页

作者： Sengupta, Anirban Anshul, Aditya Department of Computer Science and Engineering Indian Institute of Technology (IIT) Indore 453552 India

The increasing usage of image processing applications in modern technological environments is driven by their ability to enhance visual quality in diverse applications, from social media to medical imaging. The design of these cores as dedicated application-specific hardware intellectual property (IP) must ensure optimized performance and security. It is crucial to optimize these designs while enhancing their security to create a robust, secure design flow that can provide security against hardware security threats like IP piracy and fraudulent ownership claim. The proposed methodology introduces a low-cost security framework for application-specific hardware systems using an HLS-aided fusion watermarking framework and particle swarm optimization-driven design space exploration (PSO-DSE) process. The proposed approach demonstrates the generation of a multi-stage encryption-based signature along with the IP seller’s fingerprint biometric-based signature, which is further fused to generate the final watermark signature. It allows the embedding of a greater number of secret watermark constraints offering lower probability of coincidence and higher tamper tolerance without incurring design overhead, compared to prior works. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Fingerprint biometric Hardware security HLS Image processing applications IP piracy

来源：评论

学校读者我要写书评

暂无评论

A Low Resource Multi-lingual Simultaneous Script Identification and Text Recognition Model

引用

SN computer Science 2024年第6期5卷 740页

作者： Mukherjee, Jayati Roy, Utpal Computer Science and Engineering Academy of Technology Department of computer and system sciences Visva-bharati

In this paper, we have proposed a multi-task learning model for multi-lingual Optical Character Recognition. Our model does the script identification and text recognition simultaneously of offline machine printed documents. We have extracted the spatial and temporal features of a line image by the combination of several CNN and BLSTM layers. The feature is shared between the script identification and text recognition modules. Fully connected layer and softmax identify the script. The identified script works as a case selector for the text recognizer which is a CTC layer. Finally, the text is identified by the text recognizer. The model is applied to two public datasets: ISIDDI, RETAS containing Bengali degraded, and English pages. We have created a dataset of Devnagari/Hindi and Tamil scripts to test our model. The model has achieved 99.2% accuracy for script recognition. The achieved text recognition accuracy on the scripts Bengali, English, Hindi, and Tamil are respectively 91.68%, 97.07%, 95.68% and 92.27%. © The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd. 2024.

关键词： Deep learning Multi-task learning Optical character recognition

来源：评论

学校读者我要写书评

暂无评论

Priority-based Residential Demand Response for Alleviating Crowding in Distribution Systems

引用

Journal of Modern Power Systems and Clean Energy 2023年第2期11卷 502-510页

作者： Venkateswarlu Gundu Sishaj P.Simon Vemalaiah Kasi Narayana Prasad Padhy Dheeraj Kumar Khatod Department of Computer Science and Engineering Koneru Lakshmaiah Education Foundation GunturGunturIndia Department of Electrical and Electronics Engineering National Institute of Technology TiruchirappalliTiruchirappalliIndia Department of Electrical Engineering(EED) Indian Institute of Technology Roorkee(IITR)RoorkeeIndia

The dynamic pricing environment offers flexibility to the consumers to reschedule their switching *** the dynamic pricing environment results in several benefits to the utilities and consumers,it also poses some *** crowding among residential customers is one of such *** scheduling of loads at low-cost intervals causes crowding among residential customers,which leads to a fall in voltage of the distribution system below its prescribed *** order to prevent crowding phenomena,this paper proposes a priority-based demand response program for local energy *** the program,past contributions made by residential houses and demand are considered as essential parameters while calculating the priority *** non-linear programming(NLP)model proposed in this study seeks to reschedule loads at low-cost intervals to alleviate crowding *** the NLP model does not guarantee global optima due to its non-convex nature,a second-order cone programming model is proposed,which captures power flow characteristics and guarantees global *** proposed formulation is solved using General Algebraic Modeling System(GAMS)software and is tested on a 12.66 kV IEEE 33-bus distribution system,which demonstrates its applicability and efficacy.

关键词： Crowding strategy demand response distribution system General Algebraic Modeling System(GAMS) local energy community second-order cone programming

来源：评论

学校读者我要写书评

暂无评论

Protecting Recommendations: Combating BIAS and Manipulation in Recommendation 8

Protecting Recommendations: Combating BIAS and Manipulation ...

引用

8th International Conference on I-SMAC (IoT in Social, Mobile, Analytics and Cloud), I-SMAC 2024

作者： Infant, P.Alwin Kumar, P.Mohan Jainish, G.R. Hindustan Institute of Technology & Sciencce Department of Computer Science and Engineering Chennai India Loyola Icam College of Engineering & Technology Department of Computer Science and Engineering Chennai India

ISBN: (纸本)9798350376425

Recommender systems assist consumers in navigating the deluge of information by helping them find services and goods. The effectiveness of recommender systems has been thoroughly examined in research currently available from a number of angles. In the research community, robustness against malicious attack is becoming a more important concern. Collaborative Filtering (CF) recommender systems have been established to accomplish this objective. However, due to the fact that they are collaborative recommender systems are open to attack. To weaken the system, attackers can introduce fictitious and biased profiles. An examination of the random attack, Bandwagon attack, Mixed Attack and Noise Injection on recommender systems is presented in this work. Numerous algorithms are susceptible to profile injection attacks. Some of the earlier approaches suffer from only looking at one aspect of user rating values while assessing the rating matrix, ignoring alternative viewpoints. First, a rating matrix in this research that takes into account both genuine and fictitious user ratings is created. Next, using user rating time as a basis, the rating matrix is examined. Next, build a model to distinguish between real and fraudulent users using a variety of classifiers, including Principal Component Analysis (PCA), K-Nearest Neighbour (KNN), and Bayesian Networks (BN). Ultimately, a number of tests are conducted, and a summary of the findings comparing different approaches' abilities to detect small-scale attacks and detect overall attacks is presented. © 2024 IEEE.

关键词： Collaborative filtering

来源：评论

学校读者我要写书评

暂无评论

A Generic, High-Performance, Compression-Aware Framework for Data Parallel DNN Training

引用

IEEE Transactions on Parallel and Distributed Systems 2023年第8期36卷 1-20页

作者： Wu, Hao Wang, Shiyi Bai, Youhui Li, Cheng Zhou, Quan Yi, Jun Yan, Feng Chen, Ruichuan Xu, Yinlong Department of Computer Science and Technology University of Science and Technology of China Hefei China Computer Science Department and Electrical and Computer Engineering Department University of Houston USA Nokia Bell Labs

Gradient compression is a promising approach to alleviating the communication bottleneck in data parallel deep neural network (DNN) training by significantly reducing the data volume of gradients for synchronization. While gradient compression is being actively adopted by the industry (e.g., Facebook and AWS), our study reveals that there are two critical but often overlooked challenges: 1) inefficient coordination between compression and communication during gradient synchronization incurs substantial overheads, and 2) developing, optimizing, and integrating gradient compression algorithms into DNN systems imposes heavy burdens on DNN practitioners, and ad-hoc compression implementations often yield surprisingly poor system performance. In this paper, we propose a compression-aware gradient synchronization architecture, CaSync, which relies on flexible composition of basic computing and communication primitives. It is general and compatible with any gradient compression algorithms and gradient synchronization strategies and enables high-performance computation-communication pipelining. We further introduce a gradient compression toolkit, CompLL, to enable efficient development and automated integration of on-GPU compression algorithms into DNN systems with little programming burden. Lastly, we build a compression-aware DNN training framework HiPress with CaSync and CompLL. HiPress is open-sourced and runs on mainstream DNN systems such as MXNet, TensorFlow, and PyTorch. Evaluation via a 16-node cluster with 128 NVIDIA V100 GPUs and a 100 Gbps network shows that HiPress improves the training speed over current compression-enabled systems (e.g., BytePS-onebit, Ring-DGC and PyTorch-PowerSGD) by 9.8%-69.5% across six popular DNN models. IEEE

关键词： Synchronization

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

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

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：