检索结果-内蒙古大学图书馆

Improved optimizer with deep learning model for emotion detection and classification

Mathematical Biosciences and engineering 2024年第7期21卷 6631-6657页

作者： Joseph, Willson C. Kathrine, G. Jaspher Willsie Shanmuganathan, Vimal Sumathi, S. Pelusi, Danilo Valencia, Xiomara Patricia Blanco Verdú, Elena Department of Computer Science and Engineering Karunya Institute of Technology and Sciences Coimbatore India Department of Computer Science and Engineering Adi Shankara Institute of Engineering and Technology Kerala India Department of Artificial Intelligence and Data Science Sri Eshwar College of Engineering Coimbatore India Sri Eshwar College of Engineering Coimbatore India Communication Sciences University of Teramo Coste Sant’agostino Campus Teramo Italy Universidad Internacional de La Rioja La Rioja Logroño Spain

Facial emotion recognition (FER) is largely utilized to analyze human emotion in order to address the needs of many real-time applications such as computer-human interfaces, emotion detection, forensics, biometrics, and human-robot collaboration. Nonetheless, existing methods are mostly unable to offer correct predictions with a minimum error rate. In this paper, an innovative facial emotion recognition framework, termed extended walrus-based deep learning with Botox feature selection network (EWDL-BFSN), was designed to accurately detect facial emotions. The main goals of the EWDL-BFSN are to identify facial emotions automatically and effectively by choosing the optimal features and adjusting the hyperparameters of the classifier. The gradient wavelet anisotropic filter (GWAF) can be used for image pre-processing in the EWDL-BFSN model. Additionally, SqueezeNet is used to extract significant features. The improved Botox optimization algorithm (IBoA) is then used to choose the best features. Lastly, FER and classification are accomplished through the use of an enhanced optimization-based kernel residual 50 (EK-ResNet50) network. Meanwhile, a nature-inspired metaheuristic, walrus optimization algorithm (WOA) is utilized to pick the hyperparameters of EK-ResNet50 network model. The EWDL-BFSN model was trained and tested with publicly available CK+ and FER-2013 datasets. The Python platform was applied for implementation, and various performance metrics such as accuracy, sensitivity, specificity, and F1-score were analyzed with state-of-the-art methods. The proposed EWDL-BFSN model acquired an overall accuracy of 99.37 and 99.25% for both CK+ and FER-2013 datasets and proved its superiority in predicting facial emotions over state-of-the-art methods. ©2024 the Author(s), licensee AIMS Press.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Adaptive Three Layer Hybrid Reconfigurable Intelligent Surface for 6G Wireless communication: Trade-Offs and Performance 8

Adaptive Three Layer Hybrid Reconfigurable Intelligent Surfa...

引用

8th IEEE Asia Pacific Conference on Wireless and Mobile, APWiMob 2023

作者： Ratul, Rashed Hasan Iqbal, Muhammad Ashraf, Tabinda Pan, Jen-Yi Wang, Yi-Han Lien, Shao-Yu Department of Electrical and Electronic Engineering Dhaka Bangladesh Department of Communication Engineering Chiayi Taiwan Department of Computer Science and Information Engineering Chiayi Taiwan Tainan Taiwan

ISBN: (纸本)9798350327861

A potential candidate technology for the development of future 6G networks has been recognized as Reconfigurable Intelligent Surface (RIS). However, due to the variation in radio link quality, traditional passive RISs only accomplish a minimal signal gain in situations with strong direct links between user equipment (UE) and base station (BS). In order to get over this fundamental restriction of smaller gain, the idea of active RISs might be a suitable solution. In contrast to current passive RIS, which simply reflects and directs signals without any additional amplification, active RISs have the ability to enhance reflected signals by the incorporation of amplifiers inside its elements. However, with additional amplifiers, apart from the relatively complex attributes of RIS-assisted arrangements, the additional energy consumption of such technologies is often disregarded. So, there might be a tradeoff between the additional energy consumption for the RIS technologies and the overall gain acquired by deploying this potential advancement. The objective of this work is to provide a primary idea of a three-layer hybrid RIS-assisted configuration that is responsive to both active and passive RIS, as well as an additional dormant or inactive state. The single RIS structure should be capable of adjusting its overall configuration in response to fluctuations in transmit power and radio link quality. Furthermore, our fabricated passive RIS-assisted structure verifies a portion of the proposed idea, with simulations highlighting its advantages over standalone passive or active RIS-assisted technologies. © 2023 IEEE.

关键词： Energy utilization

来源：评论

学校读者我要写书评

暂无评论

Intention-Interaction Graph Based Hierarchical Reasoning Networks for Human Trajectory Prediction

引用

IEEE Transactions on Multimedia 2022年 1-12页

作者： Li, Cunyan Yang, Hua Sun, Jun Department of Electronic Engineering Shanghai Jiao Tong University Shanghai China Institute of Image Communication and Network Engineering and Department of Electronic Engineering Shanghai Jiao Tong University Shanghai China

Understanding crowd motion dynamics and forecasting the future pedestrian trajectories are critical to various applications, e.g. autonomous driving and surveillance system. This task is challenging because when pedestrians plan the future paths in real crowd scenes, they will distinguish the priorities of following their predetermined destinations and responding to the motion behaviors of neighboring pedestrians. However, most of the existing methods ignore the problem of intention-interaction trade-off. In this paper, we tackle this problem by a hierarchical network, which achieves dynamically reasoning predetermined destinations and future trajectories. A novel graph structure called Intention-Interaction Graph (IIG) is designed to jointly model the self intentions and social interactions. To aggregate information in IIG, Interaction Gated Graph Attention Networks (IGGAN) consisting of a gate mechanism and an attention mechanism is proposed, thus achieving reasoning the influence degree of neighboring pedestrians and destinations. Experimental results on multiple widely used pedestrian trajectory prediction datasets, including two datasets in ETH and three datasets in UCY, demonstrate the effectiveness of the proposed model. IEEE

关键词： Trajectories

来源：评论

学校读者我要写书评

暂无评论

Intrusion detection system with walrus optimization algorithm (WOA) and BiGRU-CNN for securing IoT systems

引用

International Journal of Information Technology (Singapore) 2025年 1-8页

作者： Appathurai, Ahilan Bachan, P. Kaushik, Ruchi Tamizhselvi, A. Jodha, Rajshree Singh, Kulwant Department of Electronics and Communication Engineering PSN College of Engineering and Technology Tamil Nadu Melathediyoor India Department of Electronics and Communication Engineering GLA University Mathura India Xavier Institute of Management and Entrepreneurship Bengaluru India St. Joseph’s College of Engineering Chennai India Computer Science and Engineering Indian Institute of Information Technology (IIIT) Kota India Electronics and Communication Engineering Yadavindra Department of Engineering (YDoE) Punjabi University Guru Kashi Campus (PUGKC) District Bathinda Talwandi Sabo India

Security has become a primary issue in IoT (Internet of Things) because of its complex environment, handling multiple devices, several connection methods, and transmission of a huge volume of data. Traditional techniques don’t manage datasets or learn features optimally, which degrades attack detection accuracy. To solve the issues, an effective Intrusion Detection System (IDS) is proposed for detecting security risks in IoT systems. The framework comprises three phases, namely data pre-processing, feature selection, and classification. Pre-processing is applied to the data gathered from the public repository. The best features are then selected from the pre-processed dataset by inputting the pre-processed data into the Walrus Optimization Algorithm (WOA). It is followed by data classification using Bidirectional Gated Recurrent Unit-Convolutional Neural Network (BiGRU-CNN). This framework can classify data into five categories, such as normal (benign), DoS (Denial of Service), Shellcode (U2R), Fuzzers (R2L), and Probe attacks. The effectiveness of the proposed framework is evaluated based on metrics such as intrusion detection rate (IDR %) and network accuracy. The results are highly satisfactory, leading to network accuracy and IDR as 98.20% and 98.15%, respectively. © Bharati Vidyapeeth's institute of computer Applications and Management 2025.

关键词： BiGRU-CNN IDR IDS IoT security Network accuracy WOA

来源：评论

学校读者我要写书评

暂无评论

A weighted block cooperative sparse representation algorithm based on visual saliency dictionary

引用

CAAI Transactions on Intelligence Technology 2023年第1期8卷 235-246页

作者： Rui Chen Fei Li Ying Tong Minghu Wu Yang Jiao College of Information and Communication Engineering Nanjing Institute of TechnologyNanjingChina College of Electric Power Engineering Nanjing Institute of TechnologyNanjingChina College of Electrical and Electronic Engineering Hubei University of TechnologyWuhanChina Department of Statistics University of TorontoTorontoOntarioCanada

Unconstrained face images are interfered by many factors such as illumination,posture,expression,occlusion,age,accessories and so on,resulting in the randomness of the noise pollution implied in the original *** order to improve the sample quality,a weighted block cooperative sparse representation algorithm is proposed based on visual saliency ***,the algorithm uses the biological visual attention mechanism to quickly and accurately obtain the face salient target and constructs the visual salient ***,a block cooperation framework is presented to perform sparse coding for different local structures of human face,and the weighted regular term is introduced in the sparse representation process to enhance the identification of information hidden in the coding ***,by synthesising the sparse representation results of all visual salient block dictionaries,the global coding residual is obtained and the class label is *** experimental results on four databases,that is,AR,extended Yale B,LFW and PubFig,indicate that the combination of visual saliency dictionary,block cooperative sparse representation and weighted constraint coding can effectively enhance the accuracy of sparse representation of the samples to be tested and improve the performance of unconstrained face recognition.

关键词： cooperative sparse representation dictionary learning face recognition feature extraction noise dictionary,visual saliency

来源：评论

学校读者我要写书评

暂无评论

Sentiment Analysis with Deep Learning Methods for Performance Assessment and Comparison 3

Sentiment Analysis with Deep Learning Methods for Performanc...

引用

3rd International Conference on Image Processing and Robotics, ICIPRoB 2024

作者： Sultana, Achhiya Hasan, Mahady Shidujaman, Mohammad Premachandra, Chinthaka Independent University Bangladesh Department of Computer Science and Engineering Dhaka Bangladesh Shibaura Institute of Technology Department of Electronic Engineering Tokyo Japan

ISBN: (纸本)9798350374766

The activity of obtaining and evaluating perspectives of individuals, feelings, mindsets of others, views, and so on, toward various things including subjects, goods, and ideas is known as sentiment analysis (SA), often called sentiment mining (SM). People are producing massive quantities of thoughts and feedback regarding goods, offerings, and daily operations as a result of the quick expansion of using online applications like blogs, social media platforms, and web pages. Companies, government agencies, and institutions can collect and evaluate general population attitudes along with opinions using sentiment analysis to acquire business insight and improve decision-making. The paper represents a complete research on sentiment analysis based on DL (deep learning) approaches to give researchers an idea of the evaluation of feelings and associated disciplines. This research represents the previous studies of emotional analysis and illustrates the methodology of our work. The methodology explains data extraction, data preprocessing, text preprocessing, feature extraction, feature selection, and so on. The dataset applied in the study is an IMDb movie reviews dataset containing equal amounts of samples for training and testing. Then, we discussed sentiment analysis techniques which are Simple Neural Networks (SNN), Convolutional Neural Networks (CNN), and Recurrent Neural Networks (RNN). Using the methods, the outcome states that the simple neural network model generates an accuracy of 74.99% and a Convolutional Neural Network of 85.79%. Besides, the Recurrent Neural Network shows 86.46% which is the highest one. Furthermore, based on the results of the confusion matrix, we investigated the optimum model to attain the highest precision, recall, and F1 score. © 2024 IEEE.

关键词： Sentiment analysis

来源：评论

学校读者我要写书评

暂无评论

Effect of Substrate Selection on the Design of Inset-Fed Patch Microstrip Antennas for Wireless communication

Effect of Substrate Selection on the Design of Inset-Fed Pat...

引用

2024 Asia Pacific Conference on Innovation in Technology, APCIT 2024

作者： Muthuswamy, Jayanthi Parvathy, A. Reddy, Chanda V Sree, B. Hima Reddy, B. Dheeraj Chowdary, G. Krithi New Horizon College of Engineering Department of Electronics and Communication Engineering Bengaluru India SASTRA University School of Electrical and Electronics Engineering Thanjavur India K. S Institute of Technology Department of Computer and Communication Engineering Bengaluru India

ISBN: (纸本)9798350361537

In the field of wireless communications, the design and optimization of microstrip patch antennas play a crucial role in achieving efficient and reliable wireless connectivity. This paper is used to explore the performance characteristics of an inset-fed patch microstrip antenna employing various substrates. In this work, we investigate the impact of substrate selection on the antenna's performance in order to advance ourunderstanding and contribute to the development of more effective wireless communication systems. The effectiveness of inset-fed patch microstrip antennas using various substratematerials is examined in this proposed work. In order to fabricate antennas, a variety of substrates, including as FR-4, high-frequency dielectrics, are used. Gain, radiation patterns, and impedance matching are the main considerations. Engineers can choose the best materials and insights into how substrate features affect antenna characteristics. This proposedwork provides valuable suggestion for antenna design and optimization by investigating radiation patterns, gain, and return loss for each substrate using computer Simulation Technology (CST) Microwave studio. © 2024 IEEE.

关键词： Slot antennas

来源：评论

学校读者我要写书评

暂无评论

Optimizing City Transit: IoT and Gradient Boosting Algorithms for Accurate Bus Arrival Predictions 2

Optimizing City Transit: IoT and Gradient Boosting Algorithm...

引用

2nd IEEE International Conference on Intelligent and Innovative Technologies in Computing, Electrical and electronics, ICIITCEE 2024

作者： Shanthi, S. Maruthu Kannan, B. Giriprasad, S. Babuji, R. Sivakumar, S. Malathi, N. Care College of Engineering Department of Electronics and Communication Engineering Tamilnadu Tiruchirappalli India Sphoorthy Engineering College Department of Computer Science and Engineering Telangana Hyderabad India Srm Institute of Science and Technology Department of Electronics and Communication Engineering Tamil Nadu Chennai India R.M.K. Engineering College Department of Electronics and Communication Engineering Tamil Nadu Chennai India Department of Computer Science Tamil Nadu Perambalur India Saveetha University Saveetha School of Engineering Saveetha Institute of Medical and Technical Sciences Department of Computer Science and Engineering Tamil Nadu Chennai India

ISBN: (纸本)9798350306415

Sustainable cities and people's well-being depend on efficient and reliable public transportation networks. To improve the reliability of bus arrival forecasts, we offer a new method that uses Internet of Things (IoT) data and advanced Gradient Boosting Machine learning algorithms to combine the two. Using real-time information from IoT sensors installed on buses and at strategic nodes along their routes, it constructs a comprehensive dataset that accounts for many variables contributing to bus transit delays, such as traffic, weather, and passenger load. Then, it creates superior prediction models than conventional approaches. It uses Gradient Boosting, an effective ensemble learning methodology. The findings will lead to a more dependable and efficient public transportation system by improving the accuracy of predicted bus arrival times. It helps in the continuous struggle over urban problems with mobility and sets the path for more data-driven, efficient improvement of public transportation. © 2024 IEEE.

关键词： Bus Arrival Predictions Public Transit Efficiency Public Transportation Real-time Data Smart Cities

来源：评论

学校读者我要写书评

暂无评论

Wearable Saudi Sign Language Recognition Device Based on Neural Network 21

Wearable Saudi Sign Language Recognition Device Based on Neu...

引用

21st International Multi-Conference on Systems, Signals and Devices, SSD 2024

作者： Algahtani, Shouq Zuhairy, Rana Almarghalani, Olla Aljohani, Mawadda Elmanfaloty, Rania A. King Abdulaziz University Department of Electrical and Computer Engineering Jeddah Saudi Arabia Alexandria Higher Institute of Engineering and Technology Department of Electronics and Communication Engineering Alexandria Egypt

ISBN: (纸本)9798350374131

In Saudi Arabia, over 750,000 individuals are deaf or mute, often facing communication barriers due to limited sign language familiarity among the general population. This research introduces a wearable device leveraging neural networks to recog-nize Saudi Sign Language, aiming to bridge this communication gap. The device simultaneously processes data from both hands, integrating inputs from the non-dominant hand with those from the dominant. Three primary phases define the system: data acquisition, machine learning model development, and real-time sign recognition. Data were gathered using electromyography (EMG) and inertial measurement unit (IMU) sensors, facilitated by the Arduino IDE, with two Arduino Nano 33 BLE micro-controllers managing collection and transmission. The dataset, comprising 25 signs from four subjects with each sign repeated 20 times, informed the machine learning process. Utilizing Python, a model incorporating dual Convolutional Neural Networks for feature extraction from sensor readings and dense neural networks for training was built. Subsequent sign recognition was realized by deploying the model on Arduino via TensorFlow Lite. Impressively, this innovative system boasts a 92 % accuracy rate, marking a significant stride toward inclusive communication solutions. © 2024 IEEE.

关键词： Data acquisition

来源：评论

学校读者我要写书评

暂无评论

Prediction of Lung Infections with Deep Learning Techniques: A Systematic Review 3

Prediction of Lung Infections with Deep Learning Techniques:...

引用

3rd International Conference on Innovative Mechanisms for Industry Applications, ICIMIA 2023

作者： Thangamuthu, Vaishnavi Kamaraj, K. Hezekiah, James Deva Koresh KPR Institute of Engineering and Technology Department of Computer Science and Engineering Coimbatore India KPR Institute of Engineering and Technology Department of Electronics and Communication Engineering Coimbatore India

ISBN: (纸本)9798350343632

This research study systematically examines the application of deep learning techniques in predicting lung infections including cancer, pneumonia, tuberculosis, Covid 19 and so on. The analysis includes a range of deep learning models, including convolutional neural networks (CNNs) and various architectures like ResNet, VGG, AlexNet, and others. The review assesses the efficacy of these techniques in forecasting lung infections on their computational complexity and accuracy. This study also diversified by utilizing both chest CT images and X-ray images on the predictions. The insights derived from this systematic review contribute to a better understanding of the potential and limitations of employing deep learning in the early detection of lung infections. © 2023 IEEE.

关键词： 2019-nCoV AlexNet CNN GoogleNet LeNet ResNet SARS-CoV-) VGG

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：