检索结果-内蒙古大学图书馆

3rd International Conference on Integrated Circuits and Communication Systems, ICICACS 2025

作者： Raman, Ramakrishnan Rajan, Karthick Tlajiya, Falguni Paliwal, Manisha Pune India Chennai Institute of Technology Department of Computer Science and Engineering Chennai India Parul University Faculty of Engineering and Technology Department of Electrical Engineering Vadodara India Pune India

ISBN: (纸本)9798331508456

Semi-supervised object detection (SSOD) has emerged as a critical approach to bridge the gap between the extensive labeled data requirements of supervised methods and the abundance of unlabeled data available. However, existing SSOD techniques often suffer from two significant limitations: inconsistent performance across scales and inaccuracies in object localization. To address these challenges, this paper introduces a novel framework that integrates Multi-Scale Regularization (MSR) and Bounding Box Refinement (BBR). The proposed MSR mechanism enforces consistency in predictions across different feature map scales, ensuring robust pseudo-labeling and mitigating scale-related variations. Simultaneously, BBR employs a two-stage optimization process to enhance the precision of bounding box predictions, correcting spatial errors in initial detections. The combined framework leverages both labeled and unlabeled datasets effectively, resulting in improved detection performance. Extensive experiments conducted on benchmark datasets such as COCO and Pascal VOC demonstrate the efficacy of the proposed approach, showing significant improvements in mean Average Precision (mAP) compared to existing state-of-the-art SSOD methods. Ablation studies further highlight the individual contributions of MSR and BBR to the overall performance gains. This study establishes a new benchmark in SSOD, paving the way for more accurate and efficient object detection systems that rely on minimal labeled data while leveraging abundant unlabeled data. © 2025 IEEE.

关键词： Semi-supervised learning

来源：评论

学校读者我要写书评

暂无评论

DX2CT: Diffusion Model for 3D CT Reconstruction from Bi or Mono-planar 2D X-ray(s)

DX2CT: Diffusion Model for 3D CT Reconstruction from Bi or M...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Jeong, Yun Su Yoo, Hye Bin Chun, Il Yong Department of Electrical and Computer Engineering Sungkyunkwan University Korea Republic of Departments of Artificial Intelligence and Advanced Display Engineering Sungkyunkwan University Korea Republic of Suwon16419 Korea Republic of

ISBN: (纸本)9798350368741

Computational tomography (CT) provides high-resolution medical imaging, but it can expose patients to high radiation. X-ray scanners have low radiation exposure, but their resolutions are low. This paper proposes a new conditional diffusion model, DX2CT, that reconstructs three-dimensional (3D) CT volumes from bi or mono-planar X-ray image(s). Proposed DX2CT consists of two key components: 1) modulating feature maps extracted from two-dimensional (2D) X-ray(s) with 3D positions of CT volume using a new transformer and 2) effectively using the modulated 3D position-aware feature maps as conditions of DX2CT. In particular, the proposed transformer can provide conditions with rich information of a target CT slice to the conditional diffusion model, enabling high-quality CT reconstruction. Our experiments with the bi or mono-planar X-ray(s) benchmark datasets show that proposed DX2CT outperforms several state-of-the-art methods. Our codes and model will be available at: https://***/intyeger/DX2CT. © 2025 IEEE.

关键词： Computed tomography (CT) Diffusion models Three-dimensional (3D) reconstruction X-ray radiography

来源：评论

学校读者我要写书评

暂无评论

Machine Learning Based Pattern Recognition with HOG Features for Detection of Islanding 3

Machine Learning Based Pattern Recognition with HOG Features...

引用

3rd IEEE International Conference on electrical Power and Energy Systems, ICEPES 2024

作者： Pal, Kumaresh Goswami, Subhadip Bhuyan, Anshuman Koley, Santanu Bera, Bidisha Basu, Sarmistha National Institute of Technology Department of Electrical Engineering Jamshedpur India National Institute of Technology Department of Electrical Engineering Arunachal Pradesh India Gandhi Institute for Education & Technology Department of Electrical Engineering Bhubaneswar India Haldia Institute of Technology Department of Computer Science Engineering Haldia India Global Institute of Science and Technology Department of Computer Science and Technology Haldia India Calcutta Institute of Engineering and Management Department of Physics Kolkata India

ISBN: (纸本)9798350390728

Islanding is a significant difficulty that arises from the high integration of distributed generation sources. Islanding could potentially damage the clients and their equipment. According to the IEEE 1547 DG interconnection requirements, islanding must be detected within two seconds, and the DG must be shut down. This research introduces a new method for identifying islanding using image pattern recognition utilising a Machine Learning classification algorithm. The suggested technique utilises the Random Forest (RF) classification algorithm. Images are utilised to extract the histogram of oriented gradient (HOG) features in order to detect both non-islanding and islanding occurrences. The spectrogram images are formed from the time-series signal collected from the point of common coupling. The images are utilised to extract features from histogram of oriented gradient, which are subsequently used as feature vector inputs for training and testing the RF classifiers. The spectrogram image is derived using the rate of change of the voltage parameter. The efficacy of the RF classifier is evaluated using 5, 10, 20, and 25-fold cross-validations. The classification findings indicates that islanding detection utilising a Random Forest classifier and Histogram of Oriented Gradients (HOG) feature for image pattern recognition obtained outstanding performance with 98.75% accuracy and a detection time of 192 milliseconds. © 2024 IEEE.

关键词： Spectrographs

来源：评论

学校读者我要写书评

暂无评论

Braille–Latin conversion using memristive bidirectional associative memory neural network

引用

Journal of Ambient Intelligence and Humanized Computing 2023年第9期14卷 12511-12534页

作者： Vaidyaraman, Jayasri Thyagarajan, Abitha K. Shruthi, S. Ravi, V. School of Electronics Engineering Vellore Institute of Technology Chennai India Department of Electrical Engineering and Computer Science University of Michigan Ann ArborMI United States

Artificial neural networks (ANNs) are finding increasing use as tools to model and solve problems in almost every discipline in today’s world. The successful implementation of ANNs in software—particularly in the fields of deep learning and machine learning—has spiked an interest in designing hardware architectures that are custom-made to implement ANNs. Several categories of ANNs exist. The two-layer bidirectional associative memory (BAM) is a particular class of hetero-associative memory networks that is extremely efficient and exhibits good performance for storing and retrieving pattern pairs. The memristor is a novel hardware element that is well-suited to modelling neural synapses because it exhibits tunable resistance. In this work, in order to create a device that can perform Braille–Latin conversion, we have implemented a circuit realization of a BAM neural network. The implemented hardware BAM uses a memristor crossbar array for modelling neural synapses and a neuron circuit comprising an I-to-V converter (resistor), voltage comparator, D flip-flop, and inverter. The efficiency of the implemented hardware BAM was tested initially using 2 × 2 and 3 × 3 patterns. Upon successfully verifying the ability of the implemented BAM to store and retrieve simple pattern pairs, it was trained for a pattern-recognition application, namely mapping Braille alphabets to their Latin counterparts and vice versa. The performance of the implemented BAM network is robust even on the introduction of noise. The application can recognize the input patterns with accuracies of 100% in either direction when tested with up to 30% noise. © 2022, The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

A Simple Analysis and Discussion of Poland Longwave Broadcasting Station Antenna

A Simple Analysis and Discussion of Poland Longwave Broadcas...

引用

2024 International Symposium on Antennas and Propagation, ISAP 2024

作者： Wu, Chao Qiu, Shuang Qiu, Jinghui Denisov, Olensandr Molebny, Vasyl Department of Microwave Engineering Harbin Institute of Technology Harbin China Department of Electrical and Computer Engineering University of Macau China

ISBN: (纸本)9798350364774

With the profound and complex changes of the international situation, the longwave communication has more and more important application value and strategic significance. And longwave can realize a large range of communication and even global communication at the low end of longwave frequency because of its advantages of low propagation loss, strong diffraction ability, etc. Therefore, longwave is widely used in a lot of fields, such as sea navigation, underground communication, underwater submarine communication, and so on. As one of the most important components of the longwave communication system, the performance of longwave antennas (LWAs) play a decisive role. In recent years, the research of LWAs mainly focuses on mechanical antenna and the traditional antenna form is less studied. In this paper, the world's most famous Poland longwave broadcasting station antenna is briefly discussed and analyzed and the development direction of the traditional longwave antenna is proposed. © 2024 IEEE.

关键词： Broadcasting antennas

来源：评论

学校读者我要写书评

暂无评论

Exploring Deep Brain Stimulation Effects in a Hyperbolic-Based Computational Model of Neuronal Spiking Patterns

Exploring Deep Brain Stimulation Effects in a Hyperbolic-Bas...

引用

2023 Congress in computer Science, computer engineering, and Applied Computing, CSCE 2023

作者： Vakil, Ardavan Faezipour, Miad School of Engineering Technology Electrical & Computer Engineering Technology Purdue University West LafayetteIN United States

ISBN: (纸本)9798350327595

Deep brain stimulation (DBS) is a device that relies on methods to desynchronize adverse neurodegenerative activities within the brain. It is a well-known treatment method for people diagnosed with Parkinson's disease (PD). The adjustment of a precise definition of DBS signal parameters is critical to breaking the synchrony and burst of the increased neuronal activity, while preventing tissue or electrode damage, and also considering reducing the energy consumption for prolonging the battery life of the DBS device. In this research in progress, we study a pulse DBS waveform with delay between its cathodic and anodic parts to disrupt and desynchronize the firing/bursting patterns of Parkinsonian neuronal cell activities. We use a hyperbolic neuronal spiking pattern model to represent the signal activity of the brain. Specifically, we model a neuron infected with bursting behavior similar to that of PD, and apply DBS current to observe the effect, computationally. The expected output of applying the proposed pulse delayed DBS waveform is to break the bursting pattern and convert the bursting spike into a regular tonic spike. In addition, the promising results of pulse delayed waveforms in terms of eliciting action potential, desynchronization of the bursting behavior, and reduced energy consumption, can potentially enhance the performance of the applied DBS. © 2023 IEEE.

关键词： deep brain stimulation hyperbolic model neuronal spiking pattern Parkinson's Disease pulse delayed wave-form

来源：评论

学校读者我要写书评

暂无评论

Machine Learning-Assisted Modeling of AC Stress-Induced Bias Temperature Instability in Oxide Channel Transistors for 2T Gain Cell eDRAM

Machine Learning-Assisted Modeling of AC Stress-Induced Bias...

引用

2025 IEEE International Reliability Physics Symposium, IRPS 2025

作者： Kwak, Jungyoun Zhang, Chengyang Jeong, Gyujun Phadke, Omkar Kirtania, Sharadindu G. Lee, Junmo Datta, Suman Yu, Shimeng School of Electrical and Computer Engineering Georgia Institute of Technology AtlantaGA30332 United States

ISBN: (纸本)9798331504779

Physics-based compact modeling requires substantial work to formulate equations and fine-tune coefficients. In this work, we propose a machine learning (ML)-assisted compact model to capture alternating current (AC)-induced positive bias temperature instability (PBTI) in top-gate indium tungsten oxide (IWO) transistors, demonstrating high accuracy (R2 = 0.94, - 0.98) across various operating conditions. Aging simulations of 2T gain cell (GC) embedded DRAM (eDRAM) reveal that the write transistor is more vulnerable to AC stress, leading to data hold failures as the number of AC cycles increases. This study highlights the advantages of ML-assisted compact modeling in improving reliability predictions for emerging semiconductor devices. © 2025 IEEE.

关键词： Three dimensional integrated circuits

来源：评论

学校读者我要写书评

暂无评论

Efficient Brain Tumor Classification through Transfer Learning Models 26

Efficient Brain Tumor Classification through Transfer Learni...

引用

26th International Conference on computer and Information technology, ICCIT 2023

作者： Mehemud, Md. Yusuf Binte Kibria, Hafsa Salam, Abdus Rajshahi University of Engineering & Technology Department of Electrical & Computer Engineering Rajshahi6204 Bangladesh

ISBN: (纸本)9798350359015

One of the most dangerous diseases, brain tumors need a quick and accurate method of detection. Wide-ranging symptoms can be caused by brain tumors, and both diagnosing and treating them can be difficult. Neurologists and radiologists occasionally fail to find the brain tumor. The American Brain Tumor Association says, about 700,000 people are coping with brain tumors. Each year, more than 80,000 new cases are diagnosed. The goal of this work is to identify brain tumors from MRI images in order to prevent earlier death. Brain image classification is done using the DenseNet121 and EfficientNetB0 pretrained transfer learning model. DenseNet121 and EfficientNetB0 have been selected for brain tumor classification because of their robust performance in image analysis tasks, their pretrained knowledge from extensive datasets like ImageNet, and their computational efficiency. The experimental result shows that the proposed model performs better than the previous work model VGG16, Support Vector Machine (SVM), Discrete Wavelet Transform (DWT), Gray Level Co-occurrence Matrix(GLCM), Hidden Markov Model (HMM), and Convolution Neural Network (CNN). These previous models' classification accuracy ranges from 94% to 96%, while the suggested model where we used 80% data for training and 20% data used for testing and average accuracies are 97.09% for DenseNet121 and 98.32% for EfficientNetB0. Also, the purpose of this work to categorize 4 classes in these 3 classes for tumor and 1 class for non tumor. Additionally, the proposed model's precision, recall, and F1-score are all 97% and 98%, indicating very good performance. However, these techniques have shown a significant potential for improving the diagnosis and treatment of brain tumors by providing a quick, accurate, and non-invasive way to categorize different types of tumors. © 2023 IEEE.

关键词： Brain tumor Classification CNN DenseNet121 EfficientNetB0 F1-score MRI

来源：评论

学校读者我要写书评

暂无评论

Terahertz Spectrum of Ions in Living Organisms

Terahertz Spectrum of Ions in Living Organisms

引用

China National Conference on Terahertz Biophysics, CTB 2025

作者： Liu, Yifei Qu, Hongkang Zhu, Zhi School of Optical-Electrical and Computer Engineering University of Shanghai for Science and Technology Shanghai200093 China

ISBN: (纸本)9789819648856

The aim of this study is to investigate the vibrational modes of common ions such as sodium (Na+), potassium (K+), magnesium (Mg2+), calcium (Ca2+) and chloride (Cl−) ions in living organisms by THz spectroscopy. These ions play key roles in intra- and extracellular signaling, cell membrane potential maintenance, neural signaling, regulation of metabolic activities, and cell volume regulation. The results show that each ion displays distinctive absorption peaks, and the positions and intensities of these peaks reflect their respective vibrational modes. The absorption peaks of Na+, Cl−, and K+ are located in the lower frequency region, whereas those of Ca2+ are located in the low frequency region, and the absorption peaks of Mg2+ are more distinctive and located in the higher frequency bands in the low frequency region. By analyzing the positions, intensities and shapes of these characteristic peaks, we are able to infer the chemical environments and possible interactions of the ions, which is essential for understanding the mechanisms of ion action in living organisms. © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

关键词： Terahertz spectroscopy

来源：评论

学校读者我要写书评

暂无评论

Electrostatic tapering for efficient generation of radiation

引用

Physical Review E 2022年第2期105卷 L023201-L023201页

作者： Miron Voin Levi Schächter Department of Electrical and Computer Engineering Technion—Israel Institute of Technology Haifa 32000 Israel

We demonstrate the feasibility of electrostatic (dc) tapering such that the net average force of the electromagnetic (ac) field is compensated by a dc field, which at resonance may be interpreted as “direct” energy transfer from the dc to the ac field. This combination persists in all three components of the setup—e-gun, resonant zone, and collector, in each one playing a different role. In equilibrium, the two field components field-emit at the cathode a density modulated cylindrical beam which is accelerated along the e-gun by the dc field; the latter also focuses the e-beam. Radiation confinement perpendicular to the e-beam is ensured by an array of dielectric Bragg-mirrors and an array of metallic hollow electrodes impose synchronous bunches and wave. This double periodic structure ensures the coexistence of dc and ac fields in the same volume. Numerical simulations demonstrate the feasibility of generation of order of 1[W] power at 1 THz from a volume on a scale of few mm3 with efficiency of the order of 25%.

关键词： Electromagnetism Low-energy multiple-particle dynamics

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：