检索结果-内蒙古大学图书馆

A Comprehensive Review On Two-Stage Object Detection Algorithms

A Comprehensive Review On Two-Stage Object Detection Algorit...

2023 IEEE International Conference on Quantum Technologies, Communications, Computing, Hardware and Embedded Systems Security, iQ-CCHESS 2023

作者： Varghese, Rejin Sambath, M. Hindustan Institute of Technology and Science Department of Computer Applications Chennai India Hindustan Institute of Technology and Science School of Computing Science Chennai India

ISBN: (纸本)9798350314946

Two-stage object detection algorithms have gained significant attention in computer vision due to their robustness and accuracy in detecting objects in images. These algorithms consist of two main stages: region proposal generation and object classification/localization. The region proposal stage efficiently generates a diverse set of candidate object regions using various techniques, narrowing down the search space. In the second stage, deep convolutional neural networks extract discriminative features from the proposed regions, enabling precise object classification and localization. These algorithms have demonstrated impressive performance in complex scenes with occlusions, scale variations, and multiple objects. However, they have limitations such as slower inference speeds, increased computational complexity, and the need for additional training steps. To address these limitations, more efficient algorithms like single-stage detectors have emerged. Two-stage object detection algorithms offer a powerful framework for accurate and robust object detection, and ongoing research aims to refine and improve these algorithms, expanding the capabilities of computer vision applications. © 2023 IEEE.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Adaptive energy-aware algorithms to minimize power consumption and SLA violation in cloud computing

引用

Recent Advances in computer science and Communications 2021年第4期14卷 1008-1015页

作者： Singh, Monika Kumar, Pardeep Tyagi, Sanjay Department of Computer Science and Applications Kurukshetra University Haryana India

Objective: With the establishment of virtualized datacenters on a large scale, cuttingedge technology requires more energy to deliver the services 24*7 hours. With this expansion and accumulation of information on a massive scale on data centers, the consumption of an excessive amount of power results in high operational costs. Therefore, there is an urgent need to make the environment more adaptive and dynamic, where the overutilization and underutilization of hosts are well known to the system and active measures can be taken accordingly. To serve this purpose, an energy-efficient method, for the detection of overloaded and under-loaded hosts, has been proposed in this paper. For implementing VM migration, VM placement decision has also been taken to save energy and reduce SLA (Service Level Agreement) rate over the cloud. Methods: In the paper, a novel adaptive heuristics approach has been presented that concerns with the utilization of resources for dynamic consolidation of VMs based on the mustered data from the usage of resources by VMs, while ensuring the high level of relevancy to the SLA. After identification of under-load and overload hosts, VM placement decision has been taken in the way that takes minimum energy consumption. A minimum migration policy has been adopted in the proposed methodology to minimize execution time. The validation of the effectiveness and efficiency of the suggested approach has been performed by using real-world workload traces in the CloudSim simulator. Results: The results shows that the proposed methodology is ideal for SLA, but costs more for VM migration. Conclusion: A deep analysis must be done in existing energy efficient approaches and a new platform should be suggested to save energy in real life. © 2021 Bentham science Publishers.

关键词： Cloud computing

来源：评论

学校读者我要写书评

暂无评论

Comparative Analysis of Machine Learning Algorithms for Securing IoT Enabled Environment 3

Comparative Analysis of Machine Learning Algorithms for Secu...

引用

3rd IEEE International Conference on Computing, Communication, and Intelligent Systems, ICCCIS 2022

作者： Sagu, Amit Gill, Nasib Singh Gulia, Preeti Rani, Deepti Chahal, Ayushi Maharshi Dayanand University Department of Computer Science and Applications Rohtak India

ISBN: (纸本)9781665462006

The IoT (Internet of Things) link system, various applications, data storage like cloud, and another service area that possibly will be a fresh entry for attackers as they uninterruptedly offer services in the organization. At this time, threats to users' privacy and malware pose significant challenges to the integrity of the Internet of Things. These extortions might lead to the loss of important information, which in turn could cause a company's finances and reputation to suffer. In this paper, we have identified anomalous activity throughout the IoT ecosystem by utilizing a variety of machine learning approaches. The results from the experiment indicate that the categorization capabilities of machine learning techniques may serve as an alternative strategy for ensuring the safety of communication inside the IoT. © 2022 IEEE.

关键词： Internet of things

来源：评论

学校读者我要写书评

暂无评论

Modular irregular labelling in Network Analysis with Complete Bipartite Graph as Auto Proctoring 4

Modular irregular labelling in Network Analysis with Complet...

引用

4th International Conference on Emerging Systems and Intelligent Computing, ESIC 2024

作者： Sivakumar, S. Vidyanandini, S. Nayak, Soumya Ranjan Aluvala, Srinivas Srm Institute of Science and Technology Kattankulathur Faculty of Science and Humanities Department of Computer Applications Chennai603203 India Srm Institute of Science and Technology Kattankulathur College of Engineering and Technology Department of Mathematics Chennai603203 India Kiit Deemed to Be University School of Computer Engineering Odisha Bhubaneswar751024 India Sr University Department of Computer Science and Artificial Intelligence Telangana Warangal506371 India

ISBN: (纸本)9798350349856

Modular irregular labeling is an edge irregular labeling of a graph in which the edges and vertices are labelled with 1, 2, 3,.. k such that the weight of all vertices are different. The complete bipartite graph is whose vertices are divided into two subsets such that no edge is in the same subset. In this paper the modular irregular labeling technique is applied in the complete bipartite graph. It is also planned to develop certain theorems in work of applying the modular irregular labeling in complete bipartite graph. An application in complete bipartite graph in the field of networking is also discussed. The aim of the work is to build the theorem and to promote an application in complete bipartite graph. The application is about the autoproctor for the online examination to make work easier for the examiner to examine students. © 2024 IEEE.

关键词： Graph theory

来源：评论

学校读者我要写书评

暂无评论

Word Embedding for Bengali Language using Domain-related Corpus 6

Word Embedding for Bengali Language using Domain-related Cor...

引用

6th International Conference on Inventive Computation Technologies, ICICT 2023

作者： Bandyopadhyay, Ashutosh Nair, Jayashree Department of Computer Science and Engineering Amrita School of Computing Amrita Vishwa Vidyapeetham Amritapuri Kollam India Department of Computer Science and Applications Amrita School of Computing Amrita Vishwa Vidyapeetham Amritapuri Kollam India

ISBN: (纸本)9798350398496

Many natural language processing (NLP) tasks, including machine translation, document classification, information retrieval, news category classification, document clustering, news category clustering, and question-answering systems require the use of distributional word vector representation which is a very low dimensional vector representation of words that is called word embedding, such as contextual and non-contextual embedding. In addition to several contextual word-embedding approaches like BERT, IndicBERT, and SahojBERT (the Bengali equivalent of BERT), this work also covers numerous embedding strategies (such as Word2Vec, GloVe, and FastText) with multiple hyper-parameters. Of late, word embeddings from general corpus such as Wikipedia dump, and common crawl corpus are very famous to make word vectors. In the case of Indian languages also such type of corpus is used to make word vectors. But it suffers in giving good accuracy in general NLP tasks like Sentiment Analysis, News Category Classification, and News Category Clustering, especially for the Bengali Language. This research demonstrates, most importantly, that word embedding from a domain-related corpus enhances the quality of the embedding as those embeddings are working far better than general pre-trained word embeddings which can be used in various NLP tasks like News Category Classification, Word Similarity, and Document Clustering like News Category Clustering using various clustering techniques like KMeans, KMedoid algorithms and also obtaining better accuracy. This work proves that making word vectors or word embedding from a domain-related corpus for a particular NLP task derives better results alongside the quality of those embeddings. Throughout this work, a Deep Neural Network model was used. And the same model was used in all the tests and findings in this work. This work clearly states if word embedding can be derived from a corpus that is related to the domain on which NLP tasks will be p

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Secure Electronic Health Record Storage and Retreival Using Blockchain and Encryption for Healthcare Application

引用

SN computer science 2023年第3期4卷 300页

作者： Saif, Sohail Mondal, Nabin Biswas, Suparna Department of Computer Applications Maulana Abul Kalam Azad University of Technology West Bengal Kolkata India Department of Computer Science & Engineering Maulana Abul Kalam Azad University of Technology West Bengal Kolkata India

With the rapid advancement of digital technologies, the electronic health record is not a concept but a reality. Electronic health records (EHR) are the patients’ health information that are stored for further reference. These recorded data in the form of electronic media are known as electronic health records and contain sensitive information about the patients. To ensure security and privacy, sensitive health records must be stored and shared through a secure communication channel ensuring confidentiality, authenticity and integrity. Blockchain provides the trust between healthcare service providers and patients i.e. healthcare service receivers. Blockchain also offers decentralization operation and transparency, which makes any application built using it more secure and also protects unauthorized access. This paper presents a blockchain-based model to solve the issue and challenges of securing the health records in an efficient and secure manner. Several popular encryption algorithms such as AES, DES, RSA, Rabbit and RC4 have been considered and implemented in this work. We have experimentally analyzed the latency to securely store and retrieve EHRs using blockchain. Experimental results show that the blockchain model with the AES-256 algorithm has the least average end-to-end delay of 635.25 ms. © 2023, The Author(s), under exclusive licence to Springer Nature Singapore Pte Ltd.

关键词： Blockchain EHR Encryption Integrity Latency Privacy

来源：评论

学校读者我要写书评

暂无评论

MITM Attack Based Detection and Prevention for ARP Poisoning in Wireless Network Environment

MITM Attack Based Detection and Prevention for ARP Poisoning...

引用

International Conference on computer and Communications Technologies (ICCCT)

作者： J. Chandru C. Bagyalakshmi Department of Computer Applications Hindustan Institute of Technology and Science

ISBN: (数字)9798331537579

ISBN: (纸本)9798331537586

In a wireless environment, a Man-in-the-Middle (MITM) attack takes place whilst a threat actor intercepts communication between two devices, generally a user's device and a Wi-Fi access point. At a point, an attacker can eavesdrop, inj ect malware, or steal delicate information. This study focuses on address resolution protocol (ARP) poisoning. An ARP is a type of cyberattack in which a threat actor impersonates someone else and use local area network (LAN) to transmit fictitious ARP (Address Resolution Protocol) packets in an attempt to create a parallel connection to their own MAC (Media Access Control) address with the IP address of a legitimate device in the access point. These lead to intercepting data intended for a legitimate device, stealing sensitive information, or launching further attacks. So, with the help of Kali Linux using a virtual box the Arp poisoning attack is initially tested using Ettercap for ARP poisoning, followed by the implementation of the MR ARP technique to detect and prevent anomalies. The tools utilized include Wireshark for network analysis, Ettercap for testing ARP poisoning, and a Python script for detection and prevention. Further enhancements may include the addition of a graphical user interface (GUI), Automated Blocking Mechanism, integration with intrusion detection systems (IDS), or extended support for IPv6.

关键词： Wireless sensor networks Protocols Prevention and mitigation Wireless networks Linux Media Access Control IP networks Local area networks Wireless fidelity Graphical user interfaces

来源：评论

学校读者我要写书评

暂无评论

A Hybrid Watermarking Scheme Using DWT and Haar Transform for Image Authentication 3rd

A Hybrid Watermarking Scheme Using DWT and Haar Transform fo...

引用

3rd International Conference on Advanced Computing and applications, ICACA 2024

作者： Jana, Sachchidananda Pal, Pabitra Mahadani, Asim Kumar Giri, Debasis Department of Computer Science & Applications Prabhat Kumar College Contai West Bengal Purba Medinipur721404 India Department of Computer Applications Maulana Abul Kalam Azad University of Technology West Bengal Nadia741249 India Department of Computer Science and Engineering Bankura Unnayani Institute of Engineering West Bengal Bankura India Department of Information Technology Maulana Abul Kalam Azad University of Technology West Bengal Nadia741249 India

ISBN: (纸本)9789819747986

The safe transfer and preservation of medical images is crucial in the age of digital healthcare. Watermarking medical images is an essential method of adding patient data, hospital logos, and other pertinent information to these images without compromising their confidentiality, integrity, or validity. To overcome the aforementioned difficulties, this abstract offers a unique and reliable medical image watermarking system that makes use of Particle Swarm Optimization (PSO), Discrete Wavelet Transform (DWT), and Singular Value Decomposition (SVD) approaches. There are three main phases to the suggested plan. The medical image is first broken down into its frequency components using DWT, which makes watermark embedding more effective. Subsequently, the PSO method is utilized to maximize the choice of image segments for incorporation, guaranteeing that the watermark remains undetectable and robust against diverse attacks. In the last step, the watermark’s strength is increased by using the SVD approach to embed it in the singular values of certain image blocks. Comprehensive tests are conducted to assess the efficacy of the suggested approach, taking into account performance measures like the Structural Similarity Index (SSIM) and Peak Signal-to-Noise Ratio (PSNR), as well as resistance against typical assaults such as noise addition, compression, and cropping. The testing findings show that the method effectively embeds the watermark while maintaining image quality and making sure it is robust to attack. The suggested method is appropriate for clinical and telemedicine applications as it also protects the confidentiality and integrity of patient data. By advancing the development of effective and safe medical image watermarking methods, this research helps to protect the privacy and veracity of critical healthcare data. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2024.

关键词： Discrete wavelet transforms

来源：评论

学校读者我要写书评

暂无评论

Efficient Deepfake Detection Using Convolutional Neural Network

Efficient Deepfake Detection Using Convolutional Neural Netw...

引用

2025 International Conference on Intelligent Control, Computing and Communications, IC3 2025

作者： Samal, Debasish Agrawal, Prateek Madaan, Vishu Goyal, Vishal School of Computer Applications Lovely Professional University Punjab India School of Computer Science & Engineering Lovely Professional University Punjab India Shree Guru Gobind Singh Tricentenary University Haryana Gurugram India Inti International University Nilai Malaysia University Punjabi University Department of Computer Science Punjab India

ISBN: (纸本)9798331527495

In recent times, generative AI has advanced quickly, and Generative Adversarial Network (GAN) has been used to create realistic-looking fake multimedia. It is still quite difficult to identify these Artificial Intelligence (AI) generated fake photos. Therefore, the proliferation of fake media incites fear in social groups and can harm a person's or community's reputation by influencing public attitudes and perceptions about them. The Convolution Neural Network (CNN) is a useful weapon for combating deepfakes. While the majority of approaches use image datasets with a few images and pre-trained models to demonstrate the fake detection accuracy, this paper presents an efficient CNN architecture, which is trained and tested on a balanced dataset of over 1,40,000 images namely 140k Real and Fake Faces dataset available on Kaggle, comprising 1,00,000 training images and 40,000 test and validation images combined. The proposed CNN model's learning rate is increased by applying an adam optimizer and sparse-categorical cross entropy. With a validation accuracy of 95.40%, a testing accuracy of 95.18% and an Area Under Curve (AUC) score of 0.98, the specially trained CNN model establishes a new standard for the identification and categorization of deepfake images. © 2025 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Machine Learning-Based Mortality Prediction of COVID-19 Patients 1

引用

10th International Conference on Innovations in computer science and Engineering, ICICSE 2022

作者： Ani, R. Deepa, O.S. Arundhathi, M. Darsana, J. Department of Computer Science and Applications Amrita Vishwa Vidyapeetham Amritapuri India Department of Mathematics Amrita Vishwa Vidyapeetham Coimbatore India

ISBN: (数字)9789811974557

ISBN: (纸本)9789811974540

Coronavirus Disease 2019 (COVID-19) has impacted our world. Risk classification during hospitalization is essential for medical planning and allocation of resources, and hence for lowering death rates, especially in underdeveloped countries. Many characteristics related to patients that influence illness severity, such as pre-existing comorbidities, can be employed to improve this prediction. Finding a biomarker which helps to identify individuals who need prompt treatment and determine their mortality risk has thus become a pressing yet difficult task. There is no advanced tool available for this. As a result, our research aims to develop a ML-based predictive model as well as a decision support system that can predict mortality based on clinical and health characteristics. We used different ML algorithms such as Random Forest, Support Vector Machines, XGBoost Classifier, and Logistic Regression for survival rate prediction in covid patients. Two feature selection methods, Information Gain and SVM-RFE were also incorporated. Accuracy comparison on 3 different datasets using the ML algorithms was done and the important biomarkers that helps in mortality prediction were identified. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： COVID-19

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：