检索结果-内蒙古大学图书馆

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： He, Zhimin Qian, Jiangbo Yan, Diqun Wang, Chong Xin, Yu Ningbo University Faculty of Electrical Engineering and Computer Science Ningbo315211 China

ISBN: (纸本)9781728163277

Re-identification (Re-ID) technology is important for wildlife conservation and intelligent farm management. With the development of deep learning, the performance of animal ReID based on computer vision has been improved. However, variations in animal pose push a negative impact on recognition performance. In this paper, a Multi-pose Feature Fusion Network (MPFNet) is proposed to improve the performance of the Re-ID. First, we construct three pose modules for the three postures, that is, standing, sitting, and lying, respectively. In each pose module, there are two parallel branches, one is a global branch for extracting global features, and the other is a local branch for extracting local features. In addition, to obtain more effective feature representations, we weighted fusion for the global branching of the three pose modules. We validate the efficiency of MPFNet on both the self-built MPDD dog dataset and the public ATRW Amur Tiger dataset. Experimental results show that MPFNet can obtain better recognition performance than other state-of-the-art Re-ID methods. The source of code will be public available at https://***/hezhimin7028/MPFNet. © 2023 IEEE.

关键词： Animal re-identification computer vision Feature fusion Posture diversity

来源：评论

学校读者我要写书评

暂无评论

A Single-Source Surface Integral Methods for Analysis of Dielectric Objects

A Single-Source Surface Integral Methods for Analysis of Die...

引用

2023 International Applied Computational Electromagnetics Society Symposium, ACES-China 2023

作者： Fu, Zhengxuan Shao, Hanru Ningbo University Faculty of Electrical Engineering and Computer Science Ningbo315211 China

ISBN: (纸本)9781733509657

Traditional surface integral methods require both of equivalent electric current and magnetic current for calculating the scattering of dielectric objects. This paper proposes a method that needs a single current, which is using the differential surface admittance operator. Compared to the Poggio-Miller-Chang-Harrington-Wu-Tsai (PMCHWT) method, the number of unknowns is only half of the PMCHWT formulation, and it requires only a quarter of the memory consumption of the PMCHWT equation. In terms of numerical verification, the method has been improved with BC (Buffa-Christiansen) basis functions and the accuracy has been verified. © 2023 Applied Computational Electromagnetics Society (ACES).

关键词： Numerical methods

来源：评论

学校读者我要写书评

暂无评论

Hyperparameter Optimization of Graph Neural Networks for mRNA Degradation Prediction 46

Hyperparameter Optimization of Graph Neural Networks for mRN...

引用

46th ICT and Electronics Convention, MIPRO 2023

作者： Vodilovska, Viktorija Gievska, Sonja Ivanoska, Ilinka Ss Cyril and Methodius University Faculty of Computer Science and Engineering Skopje Macedonia

ISBN: (纸本)9789532331042

Graph Neural Networks (GNN) emerged as increasingly attractive deep learning models for complex data, making them extremely useful in biochemical and pharmaceutical domains. However, building a good-performing GNN requires lots of parameter choices and Hyperparameter optimization (HPO) can aid in exploring solutions. This study presents a comparative analysis of several strategies for Hyperparameter optimization of GNNs. The explored optimization techniques include complex algorithms such as the bio-inspired Genetic Algorithm, Particle Swarm Optimization, and Artificial Bee Colony. In addition, Hill Climb and Simulated Annealing as well as the commonly used methods Random Search and Bayesian Search have also been covered. The proposed optimization algorithms have been evaluated on improving the performance of the GNN architectures developed for predicting mRNA degradation. The Stanford OpenVaccine dataset for mRNA degradation prediction has been used for training and testing the predictive models. Finding mRNA molecules with low degradation rates is important in development of mRNA vaccines for diseases such as COVID-19 and we hope to benefit research on ML in this domain. According to the analysis's findings, Simulated Annealing algorithm outperforms other algorithms on both architectures. Furthermore, population based algorithms like Particle Swarm optimization show promising results, with certain limitations related to the complexity of the algorithms which encourages further exploration of the subject. © 2023 MIPRO Croatian Society.

关键词： Genetic algorithms

来源：评论

学校读者我要写书评

暂无评论

Wideband Matched Filtering Antennas with Absorptive Resonators

Wideband Matched Filtering Antennas with Absorptive Resonato...

引用

2023 International Applied Computational Electromagnetics Society Symposium, ACES-China 2023

作者： Cui, Jiaqian Zhu, Lei University of Macau Faculty of Science and Technology Department of Electrical and Computer Engineering China

ISBN: (纸本)9781733509657

This paper presents a design method of filtering antennas with wide matching bands to realize the integration of multiple functions such as absorbing, filtering, and radiating, and protect the system as well. In our design, an absorptive stub is employed to absorb the out-of-band signals. As such, a cascaded topology with lossy stubs, lossless stubs, and the resonant antenna is proposed to acquire the absorptive filtering antenna. The characteristic of the absorptive stub, and the response of the equivalent circuit are specifically analyzed here. Finally, an absorptive filtering slot antenna is implemented to prove that the method can be used to achieve a matching bandwidth far beyond that of a conventional resonant antenna. © 2023 Applied Computational Electromagnetics Society (ACES).

关键词： Topology

来源：评论

学校读者我要写书评

暂无评论

Improvised Prediction of Rainfall using Random Forest Classifier over Taiwan 4

Improvised Prediction of Rainfall using Random Forest Classi...

引用

4th International Conference on Soft Computing for Security Applications, ICSCSA 2024

作者： Puttanapura, Jagadeshwari Monika Singh, T. Kishor Kumar Reddy, C. Doss, Srinath Stanley College of Engineering & Technology for Women Department of Computer Science and Engineering Hyderabad India Botho University Faculty of Engineering and Technology Botswana

ISBN: (纸本)9798331515720

Inadequate rainfall causes reduced harvesting, reduced availability of water, economic losses, environmental degradation, public health problems, decreased power generation, social unrest, and infrastructure destruction. Improved Rainfall prediction helps in better management of farming, resource allocation, scheduling, and disaster prevention. This leads to enhanced farming strategies, as well as more efficient water resource distribution and disaster preparedness. The dataset contains six weather variables including humidity, temperature, pressure, wind direction and speed. Numerous techniques, such as long short-Term memory (LSTM) model, bidirectional long short-Term memory (Bi-LSTM), Deep Learning Neural Networks (DLNN), Random Forest (RF), Artificial Neural Networks (ANN), K-Nearest Neighbor, Support Vector Machines (SVM), Logistic Regression were compared. The RF model's precision and accuracy were 86% and 96% respectively. In comparison with other models' performance, RF had the best outcomes. The results demonstrate that the RF algorithm can supply dependable meteorological forecasts. Such dependability is of use to farmers and other disaster response stakeholders, since it enhances planning and preparedness efforts through provision of reliable predictions. © 2024 IEEE.

关键词： Rain

来源：评论

学校读者我要写书评

暂无评论

Modified MobileNetV2 with Convolutional Block Attention Module for Facial Expression Recognition

Modified MobileNetV2 with Convolutional Block Attention Modu...

引用

2023 International Workshop on Artificial Intelligence and Image Processing, IWAIIP 2023

作者： Nafi'an, Muh. Hafidh Ilmi Bachtiar, Fitra Abdurrachman Setiawan, Budi Darma Brawijaya University Computer Science Faculty Informatics Engineering Department Malang Indonesia

ISBN: (纸本)9798350382914

Facial emotion recognition is one of the artificial intelligence implementations used to recognize emotions based on data learned by computers. Unlike humans, who can recognize a person's emotions directly, computers need several trainings procedures conducted by humans to be able to recognize a person's emotions. Previous research has proposed various methods with deep learning and traditional machine learning approaches to classify emotions based on faces. Some studies obtained relatively high accuracy, but on evaluation by cross-validation, the results were much lower than the accuracy obtained. Therefore, this study proposes an approach using a modified MobileNetV2 deep learning architecture in the residual layer by adding a Convolutional Block Attention Module (CBAM) to improve accuracy and data generalization. This experiment uses the Karolinska Directional Emotional Faces (KDEF) public benchmark dataset. Preprocessing is done with image augmentation to enrich the training data by resizing, center cropping, random horizontal flipping, random affine, and random rotation. The proposed method is evaluated by 5-fold cross-validation and compared with MobileNetV2 without modification and state-of-the-art methods. Experimental results show that the proposed model outperform all comparative methods by achieving a 5 -fold cross-validation accuracy of 94.387%. © 2023 IEEE.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

Optimizing Power Consumption for Business Networks through Cooperative Communication in Mixed Wireless Environments

Optimizing Power Consumption for Business Networks through C...

引用

2024 International Conference on Knowledge engineering and Communication Systems, ICKECS 2024

作者： Raman, Ramakrishnan Kumar, Vikram Sanjay, C.P. Rabadiya, Dhaval Patre, Smruti Meenakshi, R. India SRV Media Pvt. Ltd. India India Parul Institute of Technology Parul University Computer Science and Engineering Faculty of Engineering and Technology India India Chennai Institute of Technology Computer Science and Engineering India

ISBN: (纸本)9798350359688

In the evolving landscape of business networks, the imperative to optimize power consumption has become paramount, particularly within the context of mixed wireless environments. This study introduces a novel approach to enhancing energy efficiency through cooperative communication strategies, tailored for heterogeneous networks that comprise a mix of cellular, Wi-Fi, and other wireless technologies. The core proposition of this research lies in leveraging the inherent diversity and redundancy in mixed wireless environments to facilitate energy-efficient communication. By employing advanced cooperative communication techniques, such as relay-assisted transmission and network coding, the proposed method significantly reduces the power required for data transmission and reception across various network segments. The methodology encompasses a comprehensive framework that models the network as a unified system, enabling the identification and exploitation of opportunities for power savings without compromising communication reliability or throughput. A set of sophisticated algorithms underpins the proposed framework, ensuring optimal decision-making regarding routing, resource allocation, and mode selection in real-time. The effectiveness of the proposed approach is validated through rigorous simulation studies, which demonstrate substantial improvements in power efficiency compared to conventional communication strategies. The results highlight the potential of cooperative communication in achieving sustainable and energy-efficient business networks, paving the way for greener and more cost-effective operations in mixed wireless environments. © 2024 IEEE.

关键词： Business Networks Cooperative Communication Energy Efficiency Mixed Wireless Environments Resource Allocation

来源：评论

学校读者我要写书评

暂无评论

ML Based Intrusion Detection Scheme for various types of attacks in a WSN using C4.5 and CART classifiers

引用

Materials Today: Proceedings 2023年 80卷 3769-3776页

作者： Gite, Pratik Chouhan, Kuldeep Murali Krishna, K. Kumar Nayak, Chinmaya Soni, Mukesh Shrivastava, Amit Department of Computer Science and Engineering IPSA-IES Indore AB Road MP Indore India Department of Computer Science and Engineering I.T.S Engineering College Greater Noida India Department of Electronics and Communication Engineering Vignan's Institute of Information Technology Duvvada Andhra Pradesh Visakhapatnam India Faculty of Emerging Technology Sri Sri University Odisha751001 India Department of Computer Science & Engineering Jagran Lakecity University Bhopal India Department of Computer Science Engineering Swami Vivekanand College of Engineering Khandwa Road Near Tolnaka Vivekanand Knowledge City Indore India

A key component for preserving a WSN's integrity is designing an intrusion detection system (IDS). This article covers several sorts of security threats that may occur in a WSN. It proposes a detection of malicious nodes in a WSN. This system identifies four common assaults (i.e., black hole, wormhole, gray hole, and DDoS attacks) and is based on a Base Station machine learning (ML) algorithm (BS). The suggested ML algorithm continually analyzes data patterns from each node. Based on this technique, BS recognizes the network's harmful behavior and provides notifications to neighbor nodes to prevent attacker. First, the various assaults are examined and their characteristics are derived in terms of network parameters. Furthermore, the gathered pattern trains an ML algorithm. Then, the attacker node is effectively and properly classified within the BS. The NS2 simulator simulated the needed secure WSN. The experimental findings revealed excellent accuracy of the suggested method for attack detection. This precision can enhance network performance in terms of energy usage and packet delivery (PDR). © 2021

关键词： Base stations

来源：评论

学校读者我要写书评

暂无评论

Optimizing Heartbeat Classification using Bézier Interpolation 46

Optimizing Heartbeat Classification using Bézier Interpolat...

引用

46th ICT and Electronics Convention, MIPRO 2023

作者： Gusev, M. Petrovski, N. Tonkovikj, L. Sts Cyril and Methodius University in Skopje Faculty of Computer Science and Engineering Macedonia

ISBN: (纸本)9789532331042

Analog to digital conversion of electrocardio-grams depends on the sampling frequency influencing the determination of a proper heartbeat location and precision of further digital processing. We set a research question to find the optimal number of interpolation points to reduce the mistakes in the similarity check of heartbeats and classify ventricular beats. In addition, another research question aims at finding the optimal number of interpolation points applying Bézier interpolation to reveal the optimal performance/cost ratio. Final research question is to find the sampling frequency that will reveal optimal performance in classification of ventricular beats. The experiments evaluate all neighbouring pairs of heartbeats from the standard benchmark MIT-BIH arrhythmia dataset resampled to a 125 Hz sampling frequency. The results show that even one more interpolation point, which corresponds to a sampling frequency of 250 Hz, will increase the performance versus the original 360 Hz sampling frequency. At the same time, the optimal is interpolation with additional five or seven points corresponding to 750 Hz, and 1000 Hz respectively. We found that a threshold value of 34 reveals the optimal performance to conclude a change between ventricular heartbeats and others, even in a 10-bit precision of the analog-digital conversion. The processing time and performance/cost-benefit analysis show that one interpolation point is the most beneficial. © 2023 MIPRO Croatian Society.

关键词： Electrocardiograms

来源：评论

学校读者我要写书评

暂无评论

Comprehensive multimodal approach for Parkinson’s disease classification using artificial intelligence: insights and model explainability

引用

Soft Computing 2025年第3期29卷 1845-1877页

作者： Balaha, Hossam Magdy Hassan, Asmaa El-Sayed Ahmed, Rawan Ayman Balaha, Magdy Hassan Bioengineering Department J.B. Speed School of Engineering University of Louisville LouisvilleKY United States Computer Science and Control Systems Engineering Department Faculty of Engineering Mansoura University Mansoura Egypt Mathematics and Engineering Physics Department Faculty of Engineering Mansoura University Mansoura Egypt Department of Obstetrics and Gynecology Faculty of Medicine Tanta University Tanta Egypt

Parkinson’s disease (PD) is a debilitating neurodegenerative disorder affecting millions worldwide. Early detection is vital for effective management, yet remains challenging. In this study, we investigated four distinct datasets for PD detection. Through comprehensive experimentation employing ensemble methods and feature selection, we achieved high classification accuracies across the datasets. For the Oxford Parkinson’s Disease Detection Dataset, an accuracy of 95.67%, precision of 97.59%, recall of 84.5%, specificity of 99.32%, and F1-score of 90.57% were achieved. For the Alzheimer Parkinson Diseases 3 Class Dataset, the "Stacking" approach surpasses individual models, reaching an accuracy of 99.85%, precision of 99.81%, recall of 99.81%, specificity of 99.86%, and F1 of 99.81%. For the NewHandPD dataset, Regarding the Spiral category, The "Base-P32-384" model surpasses others with an accuracy of 97.35%, precision of 96.50%, recall of 98.57%, and F1-score of 97.53%. The collective "Stacking" approach proves highly effective regarding the Circle category, achieving 100% across all performance metrics. Regarding the Meander category, the "Base-P16-224" model achieves an accuracy of 97.35%, precision of 99.26%, recall of 95.71%, specificity of 99.19%, and F1 of 97.45%. The Mobile Device Voice Recordings at King’s College London (MDVR-KCL) dataset contains two datasets. Regarding the "SpontaneousDialogue" dataset, accuracy, BAC, precision, recall, specificity, and F1-score were computed, resulting in values of 94.03%, 92.83%, 90.78%, 100.0%, and 85.67%, respectively. Regarding the "ReadText" dataset, accuracy, BAC, precision, recall, specificity, and F1-score were computed, resulting in values of 91.89%, 90.62%, 87.5%, 100.0%, and 81.25%, respectively. Our findings highlight the efficacy of leveraging diverse data sources and advanced machine learning techniques to enhance PD detection accuracy. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germa

关键词： Neurodegenerative diseases

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：