检索结果-内蒙古大学图书馆

9th International Conference on computer and Communication engineering, ICCCE 2023

作者： Septiarini, Anindita Hamdani, Hamdani Burhandenny, Aji Ery Nur, Subhan Winarno, Edy Mulawarman University Faculty of Engineering Dept. of Informatics Samarinda Indonesia Mulawarman University Faculty of Engineering Dept. of Electrical Engineering Samarinda Indonesia Universitas Stikubank Faculty of Information Technology and Industry Semarang Indonesia

ISBN: (纸本)9798350325218

The quality of coffee beans has an important effect on the price and consumer satisfaction;therefore, it is essential in the industry to determine the quality of coffee based on the beans. The coffee industry is a huge commodity, especially Robusta coffee, which is popular among many groups. However, not everyone can differentiate the quality of coffee beans consumed because this is only known by people involved in the coffee world. Accordingly, the method to classify the quality of Robusta coffee beans was developed using machine learning. The method consists of two main processes: feature extraction and classification. Feature extraction was applied based on color, shape, and texture features;then, classification was performed using several machine-learning algorithms. The performance evaluation used cross-validation with the K-Fold of 5 and 10. The highest accuracy value achieved was 96.11%. This value resulted from color and texture features utilizing k-nearest neighbor (KNN) and artificial neural network (ANN). The resulting performance shows that the proposed method accurately classified coffee beans. © 2023 IEEE.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

EffiSegNet: Gastrointestinal Polyp Segmentation through a Pre-Trained EfficientNet-based Network with a Simplified Decoder 46

EffiSegNet: Gastrointestinal Polyp Segmentation through a Pr...

引用

46th Annual International Conference of the IEEE engineering in Medicine and Biology Society, EMBC 2024

作者： Vezakis, Ioannis A. Georgas, Konstantinos Fotiadis, Dimitrios Matsopoulos, George K. Tecreando B.V. Amsterdam Netherlands National Technical University of Athens Biomedical Engineering Laboratory School of Electrical and Computer Engineering Athens Greece University of Ioannina Dept. of Materials Science and Engineering Ioannina Greece

ISBN: (纸本)9798350371499

This work introduces EffiSegNet, a novel segmentation framework leveraging transfer learning with a pre-trained Convolutional Neural Network (CNN) classifier as its backbone. Deviating from traditional architectures with a symmetric U-shape, EffiSegNet simplifies the decoder and utilizes full-scale feature fusion to minimize computational cost and the number of parameters. We evaluated our model on the gastrointestinal polyp segmentation task using the publicly available Kvasir-SEG dataset, achieving state-of-the-art results. Specifically, the EffiSegNet-B4 network variant achieved an F 1 score of 0.9552, mean Dice (mDice) 0.9483, mean Intersection over Union (mIoU) 0.9056, Precision 0.9679, and Recall 0.9429 with a pre-trained backbone - to the best of our knowledge, the highest reported scores in the literature for this dataset. Additional training from scratch also demonstrated exceptional performance compared to previous work, achieving an F 1 score of 0.9286, mDice 0.9207, mIoU 0.8668, Precision 0.9311 and Recall 0.9262. These results underscore the importance of a well-designed encoder in image segmentation networks and the effectiveness of transfer learning approaches. © 2024 IEEE.

关键词： Semantic Segmentation

来源：评论

学校读者我要写书评

暂无评论

A non-invasive closed-loop diagnostic and therapeutic wearable device for diabetic hyperglycemia prevention: System Architecture 7

A non-invasive closed-loop diagnostic and therapeutic wearab...

引用

7th South-East Europe Design Automation, computer engineering, computer Networks and Social Media Conference, SEEDA-CECNSM 2022

作者： Fiska, Vasiliki Evangelou, Sofia Giannakeas, Nikolaos Tzallas, Alexandros T. Angelidis, Pantelis Tsipouras, Markos G. Univ. of Western Macedonia Dept. of Electrical and Computer Engineering Kozani Greece Univ. of Ioannina Dept. of Informatics and Telecommunications Arta Greece

ISBN: (数字)9798350398588

ISBN: (纸本)9798350398588

Diabetes Mellitus (DM) is a metabolic disorder affecting millions of patients worldwide. Continuous glucose monitoring (CGM) is integral to managing diabetes, and when used in conjunction with closed-loop automated medication administration systems, it can considerably lower the risks of hyperglycemia in DM patients. However, such treatment frequently necessitates the use of invasive blood glucose monitoring techniques and drug infusion. Because of its unique characteristics, sweat is considered one of the most important bio-fluids utilized in non-invasive CGM applications. In order to provide patients with effective and painless DM monitoring and treatment, the architecture of a sweat-based non-invasive closed-loop system for diabetes monitoring and treatment is extensively presented. © 2022 IEEE.

关键词： Diagnosis

来源：评论

学校读者我要写书评

暂无评论

Implementation of Modulo Multiplication with CAM-Based Massive-Parallel SIMD Matrix Core

Implementation of Modulo Multiplication with CAM-Based Massi...

引用

2023 International Technical Conference on Circuits/Systems, computers, and Communications, ITC-CSCC 2023

作者： Kageyama, Kyosuke Hamano, Hajime Kayama, Ryogo Koide, Tetsushi Kumaki, Takeshi Kindai University Dept. of Electrical Electronic and Communication Engineering Osaka Japan Ritsumeikan University Dept. of Electronic and Computer Engineering Shiga Japan Hiroshima University Hiroshima Japan

ISBN: (纸本)9798350326413

Recently, semiconductor technology has been growing rapidly, and mobile devices are required to have digital convergence. In this paper, a CAM-based massive-parallel SIMD matrix core (CAMX) is proposed as a mobile device accelerator for high-performance, programmability, and versatility. The CAMX can process repeated arithmetic and table-lookup coding operations in parallel. Mobile devices need to keep secure communication on the Internet, and the cipher techniques are wide spread for secure communication like the Advanced Encryption Standard (AES), Rivest-Shamir-Adleman (RSA), and so on. In particular, RSA encryption needs to calculate modulo multiplication. In this paper, the CAMX executes modulo multiplication for secure communication. This processing involves multiplication and division processing. This processing is implemented in the CAMX, and the CAMX can execute modulo multiplication at 1,104 clock cycles in parallel. In addition, the clock cycles of the CAMX and an ARM core are compared. From the result, the CAMX can reduce clock cycles by approximately 86 % for 1,024 parallel processing data relative to the ARM core with NEON. © 2023 IEEE.

关键词： Semiconductor device manufacture

来源：评论

学校读者我要写书评

暂无评论

Sequentially-Fed Circularly Polarized Slot Antenna Array Using Printed Ridge Gap Waveguide

Sequentially-Fed Circularly Polarized Slot Antenna Array Usi...

引用

Antennas and Propagation & USNC/URSI National Radio Science Meeting

作者： Syed M. Sifat Ahmed A. Kishk Marco A. Antoniades Dept. of Electrical Computer and Biomedical Engineering Toronto Metropolitan University Toronto ON Canada Dept. of Electrical and Computer Engineering Concordia University Montreal QC Canada

ISBN: (数字)9798350369908

ISBN: (纸本)9798350369915

A wideband sequentially-fed circularly polarized slot antenna array excited by printed ridge gap waveguide (PRGW) technology to eliminate back radiation from the slots is presented. Each slot is coupled to two orthogonal printed microstrip lines from a Y-shaped power divider that equally divides the power with a 90-degree phase difference for CP radiation. Using a sequential feeding structure significantly improves the axial ratio bandwidth of the antenna compared to the single element. The proposed antenna array operates in the Ka-band, and demonstrates a −10-dB impedance bandwidth of 20.4%, a 3-dB axial ratio bandwidth of 20%, and a peak RHCP gain of 12.3 dBic at 33 GHz.

关键词： Power dividers Surface impedance Slot antennas Surface waves Gap waveguide Bandwidth Antenna feeds

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Conventional and Tripolar EEG for High-Performance Reach-to-Grasp BCI Systems

arXiv

引用

arXiv 2024年

作者： Rabiee, Ali Ghafoori, Sima Cetera, Anna Besio, Walter Abiri, Reza Dept. of Electrical Computer and Biomedical Engineering URI Kingston United States

Objective. This study aims to enhance brain-computer interface (BCI) applications for individuals with motor impairments by comprehensively comparing the effectiveness of noninvasive tripolar EEG (tEEG) with conventional EEG technology. The focus is on accurately interpreting and decoding various grasping movements, such as power grasp and precision grasp. The goal is to determine which EEG technology is more effective in processing and translating grasp-related neural signals. Approach. The approach involved conducting an experiment on ten healthy participants who performed two distinct reach-and-grasp movements: power grasp and precision grasp, with a’no-movement’ condition serving as the baseline. Our research presents a thorough comparison between EEG and tEEG in decoding grasping movements. This comparison spans several key parameters, including signal-to-noise ratio (SNR), spatial resolution via functional connectivity, event-related potentials (ERPs), and wavelet time-frequency analysis. Additionally, our study involved extracting and analyzing statistical features from the wavelet coefficients, and both binary and multiclass classification methods were employed. Four machine learning algorithms—Random Forest (RF), Support Vector Machine (SVM), Extreme Gradient Boosting (XGBoost), and Linear Discriminant Analysis (LDA)—were used to evaluate the decoding accuracies. Main results. Our results indicated that tEEG demonstrated superior performance over conventional EEG in various aspects. This included a higher signal-to-noise ratio, enhanced spatial resolution, and more informative data in ERPs and wavelet time-frequency analysis. The use of tEEG led to notable improvements in decoding accuracy for differentiating movement types. Specifically, tEEG achieved around 90% accuracy in binary and 75.97% for multiclass classification. These results are markedly better than those from standard EEG, which recorded a maximum of 77.85% and 61.27% in similar tasks, respecti

关键词： Brain computer interface

来源：评论

学校读者我要写书评

暂无评论

Real-Time Evaluation of Human Inertial properties in Immersive Simulators for Moth type Boats

Real-Time Evaluation of Human Inertial properties in Immersi...

引用

Human-computer Interaction, Optimization and Robotic Applications (HORA), International Congress on

作者： Davide Todesca Carol Sergenti Nicola Giulietti Marco Carnevale Dept. Electrical Computer and Biomedical Engineering University of Pavia Pavia Italy

ISBN: (数字)9798350394634

ISBN: (纸本)9798350394641

The aim of this work is to present a system designed for real-time markerless evaluation of human inertial properties, to be used in a Human In The Loop (HITL) simulator of a Moth class sailboat. The method is based on human body pose estimation algorithm used for 3D skeleton reconstruction from a multi-camera acquisition. The exploitation of a markerless system enables ease of use compared to marker-based systems, and reduces signal’s disturbances compared to Inertia Measurement Units (IMUs) for body pose reconstruction. The use of multi-camera setup and the integration of YOLOv8-pose model results with image triangulation techniques reduce occlusion occurrence, measurements drift and complexity. The 3D pose obtained is then used to estimate the Center of Mass (CoM) position and the Tensor of Inertia (ToI) of the skipper. The system allows for a real-time computation of the inertial properties of the skipper, thus enabling a real-time update of the mass matrix of the numerical model running the human-in-the loop simulator. This study shows the results achievable by the system in terms of inertia estimation for different skipper’s poses, and compares the cases in which the skipper is modelled as a point mass with real-time position updates with the case in which the full inertia tensor of the skipper is considered. Results show how in contexts like those of Moth class sailboat simulators the possibility to take into account and to vary in real-time the tensor of inertia of the sailor increases the accuracy achievable in the dynamic simulation.

关键词： Tensors Three-dimensional displays Pose estimation Boats Real-time systems Human in the loop Numerical models

来源：评论

学校读者我要写书评

暂无评论

Echocardiographic Image Segmentation with Vision Transformers: A Comparative Analysis of Different Loss Functions

Echocardiographic Image Segmentation with Vision Transformer...

引用

IEEE International Workshop on Medical Measurement and Applications (MEMEA)

作者： Edoardo Bosco Giovanni Magenes Giulia Matrone Dept. of Electrical Computer and Biomedical Engineering University of Pavia Pavia Italy

ISBN: (数字)9798350307993

ISBN: (纸本)9798350308006

Deep learning (DL) algorithms have demonstrated their effectiveness in performing several automatic segmentation tasks, reaching high levels of accuracy. In particular, Vision Transformers (ViT) have recently emerged as a competing alternative to Convolutional Neural Networks (CNNs), which for many years remained state-of-the-art methods for segmentation. In numerous applications, ViTs demonstrated superior performance than CNNs, thanks to their capability to capture long-range relationships within images. Recent studies focusing on ViTs aimed to address some of the main drawbacks associated with these networks, i.e. the high computational cost and the substantial volume of data required for training and deployment. MedFormer stands out as one of the most recent ViT-based networks designed for this specific purpose. It has already been demonstrated capable of outperforming other ViTs and CNNs in segmentation tasks involving Magnetic Resonance (MRI) and computer Tomography (CT) images. This architecture is highly scalable with data size and doesn't require any pretraining. Recently, we also demonstrated its effectiveness in applications involving ultrasound images. We trained and tested it on the CAMUS echocardiographic dataset. MedFormer reached, and in some cases surpassed, the performance of CNNs, even in scenarios where other ViT-based architectures fell short. However, ViTs, and MedFormer in particular, still remain relatively underexplored in the ultrasound imaging domain. To facilitate further studies involving this type of architectures, we propose a comparison of MedFormer performance in segmenting CAMUS images using different training strategies. One of the many factors that may influence model learning is the choice of a proper loss function. In the current study, we trained MedFormer with some of most common loss functions used in semantic segmentation and we compared the results in terms of Dice score.

关键词： Training computer vision Ultrasonic imaging Statistical analysis Semantic segmentation Magnetic resonance imaging Magnetic resonance

来源：评论

学校读者我要写书评

暂无评论

Igea: a Decoder-Only Language Model for biomedical Text Generation in Italian

arXiv

引用

arXiv 2024年

作者： Buonocore, Tommaso Mario Rancati, Simone Parimbelli, Enea Dept. of Electrical Computer and Biomedical Engineering University of Pavia Pavia Italy

The development of domain-specific language models has significantly advanced natural language processing applications in various specialized fields, particularly in biomedicine. However, the focus has largely been on English-language models, leaving a gap for less-resourced languages such as Italian. This paper introduces Igea, the first decoder-only language model designed explicitly for biomedical text generation in Italian. Built on the Minerva model and continually pretrained on a diverse corpus of Italian medical texts, Igea is available in three model sizes: 350 million, 1 billion, and 3 billion parameters. The models aim to balance computational efficiency and performance, addressing the challenges of managing the peculiarities of medical terminology in Italian. We evaluate Igea using a mix of in-domain biomedical corpora and general-purpose benchmarks, highlighting its efficacy and retention of general knowledge even after the domain-specific training. This paper discusses the model's development and evaluation, providing a foundation for future advancements in Italian biomedical NLP. © 2024, CC BY-NC-ND.

关键词： Computational efficiency

来源：评论

学校读者我要写书评

暂无评论

Visual Saliency Approaches for Defect Detection in Industrial Processes: A Short Review

Visual Saliency Approaches for Defect Detection in Industria...

引用

International Conference on Emerging Technologies and Factory Automation (ETFA)

作者： Marco Porta Dept. of Electrical Computer and Biomedical Engineering University of Pavia Pavia Italy

ISBN: (数字)9798350361230

ISBN: (纸本)9798350361247

Finding defects during production is crucial in industrial manufacturing. While various detection methods exist, this paper proposes a short review on techniques based on visual saliency. Saliency approaches try to mimic how humans see and are characterized by advantages such as accuracy, speed, and ability to handle complex situations. With solutions based on deep learning becoming increasingly dominant, this review aims to provide a brief analysis of the current state of research in this field, possibly stimulating continued exploration and development of new techniques.

关键词： Deep learning COVID-19 Visualization Reviews Pandemics Production Manufacturing Manufacturing automation Defect detection Textiles

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：