检索结果-内蒙古大学图书馆

2024 International Conference on Electrical, Electronics and Computing Technologies, ICEECT 2024

作者： Thomas, Bernard Bhatt, Adityavikram Singh, Shailendra Narayan Amity University ASET Department of Computer Science and Engineering Noida India

ISBN: (纸本)9798350378092

Humans are social and emotional beings. With the rapid normalization of work-from-home and hybrid work environments, most people are having a difficult time truly expressing their emotions, which when not addressed inevitably leads to drastic mental problems. In order to address or prevent these mental problems, detection and recognition of facial emotions serve as an initial step. This paper presents a facial emotion recognition model working on the CNN (convolutional neural network) architecture, trained and tested using the FER2013 dataset, achieving an accuracy of 74% of detecting seven emotions - Happy, Neutral, Fear, Sadness, Anger, Disgust, and Surprise. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Image Recognition and Classification using CNN: A Review

Image Recognition and Classification using CNN: A Review

引用

2024 International Conference on Electrical, Electronics and Computing Technologies, ICEECT 2024

作者： Yaseen, Maroofa Singh, Shailendra Narayan Amity University Noida Department of Computer Science and Engineering ASET Uttar Pradesh India

ISBN: (纸本)9798350378092

This review paper has examined the field of foggy image detection and defogging algorithms using convolutional neural networks to handle the challenges and limitations of poor visibility during harsh weather conditions such as fog. A detailed comparison of three of the most successful fog detection algorithms - FogNet, FogCNN, and DeepFog - based on a set of performance indicators including precision, recall, F1 score, false positive rate (FPR), false negative rate (FNR), and the area under the receiver operating characteristic curve (AUC-ROC) has been conducted. From the performance evaluation, it is clear that DeepFog offers the best performance in distinguishing between foggy and non-foggy images with higher levels of accuracy. In addition, we have surveyed various imaging techniques with several methodologies that outperformed previously published defogging algorithms based on metrics such as Structural Similarity Index and Peak Signal-to-Noise Ratio. Moreover, we also consider the future development of such technology through multi-modality fusion strategies, real-time optimization, adversarial training, and human-centric applications to give an impression of the advancement of robust systems for these real-world use scenarios. © 2024 IEEE.

关键词： Convolutional neural networks

来源：评论

学校读者我要写书评

暂无评论

Generative Adversarial Networks (GANs) in computer-Generated Imagery 2

Generative Adversarial Networks (GANs) in Computer-Generated...

引用

2nd IEEE International Conference on Communication, Security and Artificial Intelligence, ICCSAI 2023

作者： Shah, Sannan Ahmad Drabu, Shahhab Hasin Khan, Murtaza Mahdi Kirmani, Dawar Bin Qayoom Singh, Shailendra Narayan Amity University Computer Science and Engineering Department Aset Noida India

ISBN: (纸本)9798350369960

In order to estimate the model using the method it is necessary to train two different models simultaneously. The first model, called the model (D) is responsible, for determining whether a particular sample is from the training data or generated by another model called G. On the hand G aims to understand the underlying distribution of the data. By training G to maximize the likelihood of D making mistakes a competitive environment is established. There are applications for G and D but one particularly notable solution emerges. D consistently evaluates any input as if it came from training data while G effectively captures the details of that training data distribution. Notably this system can be trained using known techniques like backpropagation when implementing G and D as multilayer perceptrons. This eliminates the need for inference networks or intricate Markov chains, during both training and sample generation processes. Thorough evaluations of the generated samples confirm the potential of this framework as demonstrated through testing. This innovative approach offers an efficient means to capture data distributions, which could potentially revolutionize generative modeling. © 2023 IEEE.

关键词： Deep learning Generative Adversarial Network (GAN) Generator and discriminator Image classification Image synthesis Image generation Machine learning algorithms Synthetic image generation

来源：评论

学校读者我要写书评

暂无评论

A Brief Review Particle Swarm Optimization on Intrusion Detection System 1

引用

6th International Conference on Innovative Computing and Communication, ICICC 2023

作者： Nandana, G.M. Yadav, Ashok Kumar Department of Computer Science and Engineering ASET Amity University Noida India

ISBN: (数字)9789819933150

ISBN: (纸本)9789819933143

There has been an improvement in recent years towards increase in volume for remarkable data in the field of Internet and computer networks. This brings up significant security-related challenges. In the past, a number of systems that provide host or network-level security have emerged. Many conventional security measures, such as, firewalls, spyware, authentication methods and antivirus software provide some way towards ensuring security, but they are still vulnerable to cyber-attacks and other system vulnerabilities. Some noteworthy alternatives are discovered, namely Intrusion Detection & Prevention Systems are an intriguing development, but they also have significant drawbacks, such as the inability to identify new assaults and respond to them in real time. Overtime, many researchers have proposed to develop a robust as well as intuitive infrastructure intrusion detection systems (IDS) utilising a variety of neural network models, i.e. support vector machines (SVM), rough sets, Particle Swarm Optimization (PSO) and many other versions. In this paper, we discuss all proposed models have been implemented by researchers in improving IDS performance with PSO model. Also, considering different IDS implemented, we can derive various IDS datasets. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2023.

关键词： Network security

来源：评论

学校读者我要写书评

暂无评论

Enhanced Security in MANETs Using AODV Protocol

引用

Journal of The Institution of Engineers (India): Series B 2025年第1期106卷 327-338页

作者： Gupta, Sunil Singla, Sanjay Sharma, Purushottam Chitkara University Punjab Rajpura India Department of Computer Science and Engineering UIE Chandigarh University Punjab Mohali India ASET Amity University Uttar Pradesh Noida India

This research study introduces an innovative security approach for MANETs to use with the AODV (Adhoc On-Demand Distance Vector) routing protocol, which can protect against attacks like the explicit packet-dropping assault and the implicit packet-dropping attack. The first iteration of the AODV routing protocol had no safeguards in place. As a result, it can’t defend itself against any kind of cyberattack. There are several safeguards available, therefore AODV is completely safe. Attackers in a network can be reduced using the same detection technologies that can find them. The topic was investigated further, and a catastrophic fault in all current safe routing techniques was discovered. Thus, present security measures are labor-intensive and call for an intricate key-management architecture. Using digital signatures and a hash chain, we introduce a new approach to protecting the AODV routing system in this study. This method protects itself from untrusted or malicious nodes with minimum performance hits. © The Institution of Engineers (India) 2024.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

A new framework for distributed clustering based data aggregation in WSN

A new framework for distributed clustering based data aggreg...

引用

International Conference on Artificial Intelligence, Blockchain, Computing and Security, ICABCS 2023

作者： Singh, Anuj Kumar Bhushan, Shashi Kumar, Ashish School of Computer Science and Engineering Galgotias University Greater Noida India Department of Computer Science ASET Amity University Punjab Mohali India Department of Computer Science and Engineering AKG Engineering College Ghaziabaad India

ISBN: (纸本)9781032669663

Wireless sensor networks (WSNs) are used in a variety of applications, such as industrial control systems, environmental monitoring, military applications, and structural health monitoring. Energy consumption is a key factor in the lifespan of a WSN, as all sensor nodes are energy-constrained devices. In order to improve energy efficiency, various routing protocols have been developed. These protocols aim to establish paths between sensor nodes and a data sink, while also maximizing the lifespan of the network. However, designing such protocols can be challenging due to the resource and power constraints, as well as the harsh conditions in which sensor nodes often operate. In this paper, a novel approach is proposed for improving energy efficiency in WSNs through the use of a distributed clustering-based data aggregation (DCDA) algorithm. In this approach, we use a grid structure as the basis for clustering, with each cluster having a designated cluster head (CH). Rather than transmitting data directly to the base station (BS), sensor nodes (SNs) send their data to their cluster head, which then aggregates and transmits the data to the BS. This approach can extend the network lifetime by reducing the energy required for communication between SNs. Simulations were conducted to evaluate the performance of the DCDA algorithm in a grid-based WSN. The results showed that using the specified parameters for clustering in the protocol significantly reduces the overall energy consumption of the entire network. © 2024 The Author(s).

关键词： Sensor nodes

来源：评论

学校读者我要写书评

暂无评论

A Comparative Analysis on Machine Learning Techniques for Driver Distraction Detection 5th

A Comparative Analysis on Machine Learning Techniques for Dr...

引用

5th International Conference on Recent Innovations in Computing, ICRIC 2022

作者： Srivastava, Garima Singh, Shikha Department of Computer Science and Engineering ASET Amity University Uttar Pradesh Lucknow India

ISBN: (纸本)9789811998751

Driving any transport vehicle carries a significant amount of risk. Driving, on the other hand, is a necessity that cannot be ignored or replaced with something safer. The only thing that can be done in this concern is by taking precautionary measures to make driving safer and reducing the injuries and death to almost 0. The distracted driving model was built with the aim of achieving accident-free roads, with no threat and danger of any accident. This model accomplishes this task by detecting the behavior of the driver while driving the vehicle and inform the driver of any distraction which can possibly result in an accident. For this purpose, the machine learning model is trained to categorize the real-time received images of the driver into various classes of distractions. The distracted driver model has been developed on various different models. The accuracy of the predictions made by each of the models is compared to finally achieve the model which can best suit the requirement and provide highly accurate results. After performing a validation check on all the algorithms, KNN model has best accuracy and found out that KNN over LDA had ~99% accuracy which will give near about accurate detection of drivers’ distraction. We also validated it on PCA and PCA over LDA too where PCA gave the accuracy of ~81% PCA over LDA gave ~76%, but LDA gave us the best accuracy, i.e., ~99%. The authors can assist the government in detecting drivers who engage in practices such as drunk driving, rash driving, texting, calling, eating, drinking, and many more by properly implementing this model. The laws can be implemented successfully and easily using this paradigm because no human intervention is necessary. The paper discusses about building a machine learning model for detecting distractions of the driver while driving. The paper comprehensively discusses the steps and methodology required for creating such model. Finally, results were calculated for respective algorithms and ba

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Cloud Cryptography: Mechanism of Different Encryption Standards 18

Cloud Cryptography: Mechanism of Different Encryption Standa...

引用

18th INDIAcom;11th International Conference on Computing for Sustainable Global Development, INDIACom 2024

作者： Lanke, Ravikumar Md Zubair Rahman, A.M.J. Bhardwaj, Rajat Somashekhara Reddy, D. Jain, Parth Mahesh, T.R. Dept. of Computer Science and Engineering Bangalore India Al-Ameen Engineering College Karundevanpalayam Erode India Aset Amity University Department of Computer Science and Engineering Bengaluru India

ISBN: (纸本)9789380544519

The term 'cloud cryptography' describes the application of cryptographic methods and protocols to protect communications and data in cloud computing settings. The provision of computing services, including storage, processing power, and applications, via the internet is known as cloud computing. Although cloud computing has numerous advantages, such as cost-effectiveness and scalability, it also poses security risks because sensitive data is processed and stored outside of the conventional network borders of an organization. Cloud cryptography plays a pivotal role in cloud security by safeguarding against unauthorized access, data disclosure, modification, or destruction. Keeping up with the most recent cryptographic methods and protocols is crucial to addressing new threats and weaknesses in this industry, which is always changing. Organizations can safeguard their data from unauthorized access, modification, and destruction by utilizing encryption techniques. This article explores the cloud cryptography along with its different encryption standards. © 2024 Bharati Vidyapeeth, New Delhi.

关键词： Architecture Benefits Cloud Computing Cryptography Encryption

来源：评论

学校读者我要写书评

暂无评论

Face Recognition Using Deep Neural Network with MobileNetV3-Large 1

引用

International Conference on Advances and Applications of Artificial Intelligence and Machine Learning, ICAAAIML 2022

作者： Bisht, Sakshi Singhal, Abhishek Kaushik, Charu Department of Computer Science and Engineering ASET Amity University Uttar Pradesh Noida India

ISBN: (数字)9789819959747

ISBN: (纸本)9789819959730

With the advancement of computer technology, multiple advanced techniques rely on machine vision, especially biometric systems, to play an essential part. The recording shows an image or video data with a face in it, then recognizes and analyzes the face region. Detection and recognition consist of a group of Artificial Intelligence (AI) techniques. It has a broad scope of uses and has become a thriving topic of study. In our proposed model, we employed AveragePooling2D with MobileNetV3Large. The algorithm is trained on the collected dataset following a few preprocessing processes. According to the findings, MobileNetV3 with AveragePooling2D has an accuracy rate of 91.12 percent on the training dataset and 99.79 percent on the verification dataset for 50 epochs. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd 2023.

关键词： Face recognition

来源：评论

学校读者我要写书评

暂无评论

Hardware-based Dual Booting Switch with Customizable Setup Script and Cross Platform Support 2

Hardware-based Dual Booting Switch with Customizable Setup S...

引用

2nd International Conference on Applied Artificial Intelligence and Computing, ICAAIC 2023

作者： Chaturvedi, Venkatesh Arora, Deepak Amity School of Engineering & Technolgy Lucknow Amity University Aset Department of Computer Science & Engineering Uttar Pradesh India

ISBN: (纸本)9781665456302

Many people use numerous operating systems to do a variety of tasks in today's hectic world. While using Linux for programming and coding, someone may utilize Windows for some document editing. There are no major problems faced while dual booting two operating systems aside from the fact that in order to dual boot, one has to open the boot menu every time in order to select which operating system is to be booted. This process of opening the boot menu and selecting the operating system each time is a very redundant task. Hence, the idea of a hardware-based switch that would help in switching the operating system without opening the boot menu every time was born. Initial thoughts for the switch included using a control board such as the Raspberry Pi for programming the switch. Even modification of the bootloader source code in order to create some hardware-based controller was attempted. However, after several days of research and testing, it was concluded that even though previous attempts had been made to create a dual booting switch using control boards, and were successful to an extent, those solutions were too complex for the value they provided. Hence, work was started on creating a dual booting switch without using a control board and with minimum modifications to the bootloader. © 2023 IEEE.

关键词： Linux

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：