检索结果-内蒙古大学图书馆

Fully-3D-Printed Microwave Sensor for Permittivity Measurement

IEEE Sensors Journal 2025年第12期25卷 21546-21551页

作者： Pradhan, Nrusingha Charan Koziel, Slawomir Central University of Karnataka Dept. of Electronics and Communication Engineering Karnataka India Reykjavik University Engineering Optimization and Modeling Center Reykjavik102 Iceland Gdansk University of Technology Faculty of Electronics Telecommunications and Informatics Poland

This paper presents the design and performance evaluation of a novel 3D-printed microwave sensor for dielectric characterization. The sensor utilizes a modified complementary split-ring resonator (CSRR) structure integrated within the ground plane. The sensor’s ability to measure both solids and liquids is demonstrated, with solid samples placed in proximity to the resonator and fluids injected into the embedded channel, thereby avoiding additional materials for fluidic channels. The sensor’s performance is evaluated based on changes in resonant frequency due to variations in dielectric constants. For solids, the resonant frequency shifts from 4.24 GHz to 2.66 GHz with increasing permittivity. For liquids, the frequency ranges from 4.24 GHz to 3.27 GHz for permittivity values between 1 and 80. An equivalent circuit model is developed to validate the sensor’s performance. The sensor is fabricated using the Fused Deposition Modeling (FDM) method with conductive layers applied via copper aerosol spray. The sensor’s versatility is demonstrated through testing with a range of solid substrates, such as Rogers 5880, FR4, Rogers 3006, and Rogers 6010, as well as four liquid samples—methanol, ethanol, propanol, and distilled water. The sensor effectively measures dielectric constants by monitoring shifts in resonant frequency, which vary with the permittivity of the tested samples. With an overall compact design of 30 mm × 40 mm × 1.6 mm, the sensor requires minimal liquid volume and achieves comparable sensitivity, with 4.05% for solids and 0.3% for liquids. The sensor’s resolution, accuracy, and low fabrication cost make it a promising tool for precise dielectric constant measurement. © 2001-2012 IEEE.

关键词： Permittivity measurement

来源：评论

学校读者我要写书评

暂无评论

The Usage of the BP-Layers Stereo Matching Algorithm with the EBCA Camera Set

Computer Science Research Notes

引用

Computer Science Research Notes 2023年第1-2期31卷 315-322页

作者： Kaczmarek, Adam L. Gdansk University of Technology Faculty of Electronics Telecommunications and Informatics ul. G. Narutowicza 11/12 Gdansk80-233 Poland

This paper is concerned with applying a stereo matching algorithm called BP-Layers to a set of many cameras. BP Layers is designed for obtaining disparity maps from stereo cameras. The algorithm takes advantage of convolutional natural networks. This paper presents using this algorithm with a set called Equal Baseline Camera Array. This set consists of up to five cameras with one central camera and other ones aground it. Such a set has similar advantages as a stereo camera. In particular this equipment is suitable for providing 3D vision for autonomous robots operating outdoors. The research presented in this paper shows the extent to which results of using BP Layers are improving because of using the EBCA set instead of a stereo camera. © 2023 The Author(s).

关键词： Cameras

来源：评论

学校读者我要写书评

暂无评论

Automatic Speech Liveness Detection 22

Automatic Speech Liveness Detection

引用

22nd International Conference on Emerging eLearning Technologies and Applications, ICETA 2024

作者： Haluška, Renát Kupcová, Eva Pleva, Matúš Hric, František Technical University of Košice Faculty of Electrical Engineering and Informatics Department of Electronics and Multimedia Communications Kosice Slovakia

ISBN: (纸本)9798331527716

This paper addresses the growing challenge of detecting deepfake audio, which can mimic or alter speech using advanced machine learning techniques like GAN sand RNNs. Such audio manipulation poses risks in biometric security and fraud. The study focuses on developing detection methods using acoustic features such as Mel-Frequency Cepstral Coefficients (MFCCs) and zero-crossing rate (ZCR). Two neural network models were tested, with the best achieving accuracy in identifying synthetic audio. The results highlight the need for advanced detection algorithms to keep pace with evolving deepfake technologies, enhancing biometric and digital security systems. © 2024 IEEE.

关键词： Adversarial machine learning

来源：评论

学校读者我要写书评

暂无评论

Improving the Effectiveness of Image Classification Structural Methods by Compressing the Description According to the Information Content Criterion

引用

Computers, Materials & Continua 2024年第8期80卷 3085-3106页

作者： Yousef Ibrahim Daradkeh Volodymyr Gorokhovatskyi Iryna Tvoroshenko Medien Zeghid Department of Computer Engineering and Information College of Engineering inWadi AlddawasirPrince Sattam bin Abdulaziz UniversityAl-Kharj16273Saudi Arabia Department of Informatics Kharkiv National University of Radio ElectronicsKharkiv61166Ukraine Electronics and Micro-Electronics Laboratory Faculty of SciencesUniversity of MonastirMonastir5000Tunisia

The research aims to improve the performance of image recognition methods based on a description in the form of a set of keypoint *** main focus is on increasing the speed of establishing the relevance of object and etalon descriptions while maintaining the required level of classification *** class to be recognized is represented by an infinite set of images obtained from the etalon by applying arbitrary geometric *** is proposed to reduce the descriptions for the etalon database by selecting the most significant descriptor components according to the information content *** informativeness of an etalon descriptor is estimated by the difference of the closest distances to its own and other *** developed method determines the relevance of the full description of the recognized object with the reduced description of the *** practical models of the classifier with different options for establishing the correspondence between object descriptors and etalons are *** results of the experimental modeling of the proposed methods for a database including images of museum jewelry are *** test sample is formed as a set of images from the etalon database and out of the database with the application of geometric transformations of scale and rotation in the field of *** practical problems of determining the threshold for the number of votes,based on which a classification decision is made,have been *** has revealed the practical possibility of tenfold reducing descriptions with full preservation of classification *** the descriptions by twenty times in the experiment leads to slightly decreased *** speed of the analysis increases in proportion to the degree of *** use of reduction by the informativeness criterion confirmed the possibility of obtaining the most significant subset of features for classification,which guarantees a decent

关键词： Description reduction description relevance descriptor image classification information content keypoint processing speed

来源：评论

学校读者我要写书评

暂无评论

Practical autoencoder based anomaly detection by using vector reconstruction error

引用

Cybersecurity 2023年第2期6卷 1-13页

作者： Hasan Torabi Seyedeh Leili Mirtaheri Sergio Greco Department of Electrical and Computer Engineering Faculty of EngineeringKharazmi UniversityTehranIran Department of Informatics ModelingElectronics and System EngineeringUniversity of CalabriaArcavacataItaly.

Nowadays,cloud computing provides easy access to a set of variable and configurable computing resources based on user demand through the *** computing services are available through common internet protocols and network standards.n addition to the unique benefits of cloud computing,insecure communication and attacks on cloud networks cannot be *** are several techniques for dealing with network *** this end,network anomaly detection systems are widely used as an effective countermeasure against network *** anomaly-based approach generally learns normal traffic patterns in various ways and identifies patterns of *** anomaly detection systems have gained much attention in intelligently monitoring network traffic using machine learning *** paper presents an efficient model based on autoencoders for anomaly detection in cloud computing *** autoencoder learns a basic representation of the normal data and its reconstruction with minimum ***,the reconstruction error is used as an anomaly or classification *** addition,to detecting anomaly data from normal data,the classification of anomaly types has also been *** have proposed a new approach by examining an autoencoder's anomaly detection method based on data reconstruction *** the existing autoencoder-based anomaly detection techniques that consider the reconstruction error of all input features as a single value,we assume that the reconstruction error is a *** enables our model to use the reconstruction error of every input feature as an anomaly or classification *** further propose a multi-class classification structure to classify the *** use the CIDDS-001 dataset as a commonly accepted dataset in the *** evaluations show that the performance of the proposed method has improved considerably compared to the existing ones in terms of accuracy,recall,false-positive rate,and F1-score

关键词： Cloud Practical Anomaly detection Autoencoder Reconstruction error Machine learning

来源：评论

学校读者我要写书评

暂无评论

Impact of Fe-NPs doped flux on electromigration in Sn-based solder joints of chip-sized SMD components at lower Joule heating 30

Impact of Fe-NPs doped flux on electromigration in Sn-based ...

引用

30th IEEE International Symposium for Design and Technology in Electronic Packaging, SIITME 2024

作者： Wodak, Irina Geczy, Attila Krammer, Oliver Khatibi, Golta Yakymovych, Andriy Khodabakhshi, Farzad Tafferner, Zoltan Insitute of Chemical Technologies and Analytics TU Wien Faculty of Technical Chemistry Vienna Austria Budapest University of Technology and Economics Faculty of Electronics Engineering and Informatics Department of Electronics Technology Budapest Hungary

ISBN: (纸本)9798331539511

The influence of incorporating iron nanoparticles through flux doping (0-2wt.%) in lead-free SAC305 (Sn3.5Ag0.5Cu) solder joints of chip-sized surface-mounted (SMD) components on electromigration has been investigated. To ensure high current densities throughout the experiment, track-To-pad joining angles of 135° on the anode site and 90° on the cathode site were implemented. The temperature was kept relatively low to minimize material transport attributable to thermal effects. Microstructural changes were examined through cross-sectional analysis, and push-off shear tests were performed to assess the influence of current loading on mechanical reliability. After 2000 and 4000 h of loading the samples with 4 A, joints with 0402 resistors showed copper dissolution from the tracks at the cathode sites in samples prepared with undoped flux. However, among the NP doped samples, the joints produced with 0.5 wt.% Fe-NPs showed no copper dissolution. Shear performance of joints with 0402 components deteriorated in all cases after 4000 hours of current loading. © 2024 IEEE.

关键词： Soldered joints

来源：评论

学校读者我要写书评

暂无评论

Vehicle type recognition: a case study of MobileNetV2 for an image classification task 28th

Vehicle type recognition: a case study of MobileNetV2 for an...

引用

28th International Conference on Knowledge Based and Intelligent information and Engineering Systems, KES 2024

作者： Kobiela, Dariusz Groth, Jan Hajdasz, Michal Erezman, Mateusz Gdansk University of Technology Faculty of Electronics Telecommunications and Informatics Department of Software Engineering Gabriela Narutowicza 11/12 Gdansk80-233 Poland Gdansk University of Technology Faculty of Electronics Telecommunications and Informatics Department of Algorithms and Systems Modelling Gabriela Narutowicza 11/12 Gdansk80-233 Poland Gdansk University of Technology Faculty of Electronics Telecommunications and Informatics Gabriela Narutowicza 11/12 Gdansk80-233 Poland

The goal of the research was to demonstrate the full data science lifecycle through a use case of the MobileNetv2 model for vehicle image classification task using various validation and test sets, each with different difficulty level. Diverse model variations were employed, each designed to recognize images of ground vehicles and classify them into one of five possible classes: car, truck, motorcycle, bicycle, or bus. In terms of validation accuracy, the highest results were obtained by the model trained with uniformly designed train and val sets (with data normalization and augmentation), where train set also contained validation set. This model also obtained the highest accuracy results on both test sets. The superiority of MODEL 3 BASELINE is confirmed by other metrics as well: test loss, f1-score, AUC and confusion matrices (for both test sets). Results between MODEL 1 BASELINE and MODEL 2 BASELINE differed according to the test set 1 and 2 and other metrics and it was not possible to declare the superiority of one method of datasets preparation over another (original class distribution [no data normalization and no data augmentation] versus uniformly designed [with data normalization and augmentation]). The article also presents challenges and findings - the problems, key issues, and their solutions that arose during the process of data collection and tagging, as well as the preparation and evaluation of the model. © 2024 The Authors.

关键词： Classification (of information)

来源：评论

学校读者我要写书评

暂无评论

Novel Complementary Resonator for Dielectric Characterization of Substrates Based on Permittivity and Thickness

引用

IEEE Sensors Journal 2024年第1期24卷 195-203页

作者： Haq, Tanveerul Koziel, Slawomir Engineering Optimization and Modeling Center of Reykjavik University Department of Applied Engineering Reykjavik102 Iceland Gdansk University of Technology Faculty of Electronics Telecommunications and Informatics Gdansk80-233 Poland

This article presents a novel complementary resonator featuring high sensitivity, low fabrication cost, and improved performance. The proposed structure consists of a complementary concentric square and circular ring resonator (CCSCRR) with multiple splits to enhance the inductance of the resonator. The proposed CCSCRR is coupled to a microstrip transmission line (MTL) with an impedance of $50 \Omega $ to create a high-sensitivity sensor. The lumped element equivalent circuit is employed to explain the sensor's operating principle. The geometric parameters of the CCSCRR are optimized to resonate at 15 GHz, and the optimized sensor is fabricated on 0.762-mm-thick dielectric substrate AD250 ( $\varepsilon{r}$ = 2.5 ± 0.04). Dielectric materials with relative permittivity ranging from 2.5 to 10.2 and thickness from 0.508 to 1.905 mm are employed to investigate the properties of the proposed sensor and to carry out its calibration. Based on the measured resonant frequencies of the CCSCRR sensor when loaded with different materials under test (MUTs), an inverse regression model is constructed to predict the permittivity of the MUT. Comparisons with the state-of-the-art microwave devices show that the proposed design is superior in terms of sensitivity, dielectric characterization reliability, and the applicability scope in terms of the MUT's thickness and permittivity. © 2001-2012 IEEE.

关键词： Calibration

来源：评论

学校读者我要写书评

暂无评论

Fine-Tuning GPT-J for text generation tasks in the Slovak language 22

Fine-Tuning GPT-J for text generation tasks in the Slovak la...

引用

2024 IEEE 22nd World Symposium on Applied Machine Intelligence and informatics, SAMI 2024

作者： Harahus, Maros Sokolova, Zuzana Pleva, Matus Hladek, Daniel Technical University of Košice Faculty of Electrical Engineering and Informatics Department of Electronics and Multimedia Communications Košice Slovakia

ISBN: (纸本)9798350317206

This paper provides a review of the GPT-J demonstration with a specific focus on its fine-Tuning capabilities for text generation in Slovak. Through reasoning, we explore the model's fine-Tuning in tasks such as text generation. We compare the results between different sizes of datasets and different sizes of pre-Trained models. The results clearly show that the size of the dataset has an impact on the quality of the resulting model. © 2024 IEEE.

关键词： Natural language processing systems

来源：评论

学校读者我要写书评

暂无评论

Evaluating BERT-Derived Models for Grammatical Error Detection Across Diverse Dataset Distributions 66

Evaluating BERT-Derived Models for Grammatical Error Detecti...

引用

66th International Symposium ELMAR, ELMAR 2024

ISBN: (纸本)9798350375411

This paper presents a comprehensive evaluation of three language models: RoBERTa, SlovakBERT, and BERT-Multilingual, using datasets of varying sizes (ranging from 50,000 to 1,000,000 sentences) to assess their performance in natural language processing tasks in the Slovak language. The evaluation focuses on four critical metrics: accuracy, precision, recall, and F1 score across different dataset distributions (70:30, 50:50, and 30:70). The results demonstrate SlovakBERT's superior performance across nearly all metrics and distributions, particularly excelling in precision and F1 scores, which suggests its robustness and reliability in handling varied data distributions effectively. RoBERTa, while showing variability, performs well in precision under conditions of data imbalance but faces challenges in maintaining consistent accuracy. BERT-Multilingual, though slightly trailing behind, still delivers solid performance, especially in precision, making it a dependable model across varied conditions. The findings underscore the effectiveness of SlovakBERT in scenarios requiring high precision and reliability, and provide insights into the strengths and limitations of each model under different testing conditions. © 2024 IEEE.

关键词： Natural language processing systems

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：