IEEE Access最新文献_第10页

Wafer Defect Classification Algorithm With Label Embedding Using Contrastive Learning 基于对比学习的标签嵌入晶圆缺陷分类算法

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-01-09 DOI: 10.1109/ACCESS.2025.3527491

Jeongjoon Hwang;Somi Ha;Dohyun Kim

{"title":"Wafer Defect Classification Algorithm With Label Embedding Using Contrastive Learning","authors":"Jeongjoon Hwang;Somi Ha;Dohyun Kim","doi":"10.1109/ACCESS.2025.3527491","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3527491","url":null,"abstract":"Classifying wafer defects in the wafer manufacturing process is increasingly critical for ensuring high-quality production, optimizing processes, and reducing costs. Most existing methods for wafer map defect classification primarily rely on images alone for model training and prediction. However, these approaches often lack interpretability, which can hinder process improvement and problem-solving efforts. In other words, existing methods only calculate the probability of a specific image belonging to each class, making it difficult to visually judge why the image belongs to a particular class. Additionally, these methods make it challenging to assess the distance of new images from each class. Furthermore, it is difficult to obtain representative images of each class. To address these limitations, we propose a novel approach for wafer defect classification using contrastive learning with label embedding. The proposed method aims to map label information and wafer defect images into a shared latent space through contrastive learning using label embedding. This not only facilitates defect class prediction from images but also enhances interpretability by visualizing relationships between images and defects (labels) and providing representative defect images. Moreover, compared to previous methods, our approach demonstrates better classification performance and computational efficiency, even in situations with imbalanced labels. This method also shows significant potential in identifying unseen defects not defined in the original classification tasks. Consequently, the proposed approach extends its applicability beyond wafer map defect patterns, showing promising potential for use in various domains.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"9708-9717"},"PeriodicalIF":3.4,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10835094","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993202","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Design and Implementation of Multiple Ring Oscillator-Based TRNG Architecture by Using ADPLL 基于ADPLL的多环振荡器TRNG结构设计与实现

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-01-09 DOI: 10.1109/ACCESS.2025.3527507

Huirem Bharat Meitei;Manoj Kumar

{"title":"Design and Implementation of Multiple Ring Oscillator-Based TRNG Architecture by Using ADPLL","authors":"Huirem Bharat Meitei;Manoj Kumar","doi":"10.1109/ACCESS.2025.3527507","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3527507","url":null,"abstract":"A new technique for generating true random numbers by using the ADPLL (All Digital Phase Locked Loop)-based multiple ring oscillator TRNG (MURO-TRNG) is discussed in this paper. The proposed ADPLL-based MURO-TRNG contains 10 ring oscillators, 1 conventional ADPLL, 11 sampling DFFs, 1 XOR gate, and an XOR corrector-based post-processing circuit. Ring oscillators are the entropy sources for the proposed MURO-TRNG architecture, and they are constructed by ADPLL with different frequencies. A new DCO(Digital Controlled Oscillator) constructed by using 9 NOR gates and 1 DFF is designed for constructing ADPLL-based ring oscillator circuits. Conventional ADPLL operates at 3 different reference frequencies to sample the raw random bits and to provide a clock for the post-processing circuit. The proposed MURO-TRNG architecture is designed using VHDL, implemented on the Artix 7, Kintex-7, and Zynq7000 FPGAs, and simulated by the Xilinx Vivado 2015.2 tool. The designed and implemented MURO-TRNG architectures consume 2-4 LUTS and 2-4 FFs. Energy consumption per bit of the generated bitstream is in the range of 4.22 nJ/bit-5.85 nJ/bit, and throughput values are in the range of 206.82 Mbps-260.07 Mbps. The NIST SP 800-22 test is conducted to validate the randomness of the generated bit stream outputs from the post-processing circuit.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"9252-9264"},"PeriodicalIF":3.4,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10835068","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993342","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Few-Shot Object Detection in Remote Sensing: Mitigating Label Inconsistencies and Navigating Category Variations 遥感中的少镜头目标检测：减轻标签不一致和导航类别变化

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-01-09 DOI: 10.1109/ACCESS.2025.3527881

Tiancheng Si;Shenyu Kong

{"title":"Few-Shot Object Detection in Remote Sensing: Mitigating Label Inconsistencies and Navigating Category Variations","authors":"Tiancheng Si;Shenyu Kong","doi":"10.1109/ACCESS.2025.3527881","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3527881","url":null,"abstract":"Over recent years, the increasing expansion of remote sensing image (RSI) datasets has made annotation tasks more challenging and labor-intensive, drawing considerable attention toward few-shot object detection (FSOD). Nevertheless, current mainstream FSOD models are primarily designed for natural images and encounter two substantial challenges when applied to RSIs. 1) Inconsistent label assignment for novel instances between pre-training and fine-tuning confuses detectors, leading to diminished generalization performance. 2) Complex scenes within RSIs result in significant category variations, comprising high inter-class similarity and large intra-class variance, which impairs classification accuracy. Against the aforementioned challenges, we propose a novel FSOD approach in RSIs, termed EC-FSOD. Specifically, our approach introduces two key modules: Ensemble Class-free RPN (ECF-RPN) and Contrastive Prototype ETF Classifier (CPEC). The preceding module, ECF-RPN, generates proposals by integrating multiple dissimilar yet cooperative Class-free RPNs that perceive the shape and location of target objects, mitigating the confusion caused by label inconsistencies. Furthermore, the subsequent CPEC module combines two submodules, namely Contrastive Prototype Learning Network (CPLN) and Simplex ETF Classifier (SEC), to obtain a set of representative class prototypes and robust discriminative feature representations, which are employed to overcome the category variations and enhance the generalization performance of novel instances. Extensive experiments have revealed that our approach achieves top-2 results on the DIOR dataset and optimal performance on the NWPU VHR-10.v2 dataset.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"8169-8186"},"PeriodicalIF":3.4,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10835074","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142976004","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Corrections to “Effect of Material Changes on the Total Harmonic Distortions Caused by Aluminum Electrolytic Capacitors” 修正“材料变化对铝电解电容器总谐波畸变的影响”

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-01-09 DOI: 10.1109/ACCESS.2024.3521142

René Kalbitz

引用次数: 0

Memory State Dynamics in BEOL FeFETs: Impact of Area Ratio on Analog Write Mechanisms and Charging BEOL效应管的记忆状态动力学：面积比对模拟写入机制和充电的影响

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-01-09 DOI: 10.1109/ACCESS.2025.3527628

Hannes Dahlberg;Oscar Kaatranen;Karl-Magnus Persson;Arto Rantala;Jacek Flak;Lars-Erik Wernersson

{"title":"Memory State Dynamics in BEOL FeFETs: Impact of Area Ratio on Analog Write Mechanisms and Charging","authors":"Hannes Dahlberg;Oscar Kaatranen;Karl-Magnus Persson;Arto Rantala;Jacek Flak;Lars-Erik Wernersson","doi":"10.1109/ACCESS.2025.3527628","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3527628","url":null,"abstract":"This work presents dynamic state writing by combining ferroelectric (FE) polarization together with charge injection (CI) on Si-based ferroelectric MOSFETs as a novel approach for non-volatile memory design. FE capacitors are non-destructively integrated in the Back-End-of-Line (BEOL) with Si MOSFETs to create FE-Metal-FETs (FeMFETs). We explore the FE/MOS area ratio (AR) as a critical design parameter, particularly in the context of dynamic writing processes, where various voltage pulse trains are applied for analog potentiation and depression of the memory state. AR significantly influences both the electric field distribution over the FE and the extent of CI from the top electrode. Constant-pulse writing schemes enable analog threshold voltage modulation by considering the AR, with reduced voltages and faster operation for smaller ARs. Retention of intermittent states written by FE polarization combined with CI is demonstrated, illustrating the stability and effectiveness of FeMFET devices and AR optimization for memory applications.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"9923-9930"},"PeriodicalIF":3.4,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10835082","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993849","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Performance Comparison of Loop Source and Grounded-Wire Source for Shallow Refined TEM Detection 浅层精细TEM探测中环路源与接地线源的性能比较

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-01-09 DOI: 10.1109/ACCESS.2025.3527870

Zhengyu Xu;Guofeng Zhao;Jiangying Peng;Xiao Ma;Wei Liu

{"title":"Performance Comparison of Loop Source and Grounded-Wire Source for Shallow Refined TEM Detection","authors":"Zhengyu Xu;Guofeng Zhao;Jiangying Peng;Xiao Ma;Wei Liu","doi":"10.1109/ACCESS.2025.3527870","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3527870","url":null,"abstract":"Loop-source transient electromagnetic method (TEM) and grounded-wire-source TEM are two of the most popular methods in geological electromagnetic detection. Advances in power electronics have made it possible to achieve high magnetic moments by transmitting large current pulses through small excitation coils, enabling the use of independent loop sources as magnetic dipoles in active detection techniques. The independent loop source, by overcoming the attenuation of the magnetic field caused by distance, becomes an efficient and low-cost tool for shallow detection. Both the independent loop source and the grounded-wire source share similarities in detection principles and scanning modes. This paper discusses the merits and shortcomings of the loop source and grounded-wire source for shallow, multiscale refined detection. The three-dimensional model is established and the electromagnetic response characteristics are compared by COMSOL finite element simulation software. Additionally, the influence of anomalous burial depth, offset distance, and measurement height on the detection results of these two near-source methods are thoroughly discussed. Finally, it is explained by the field measurement results. The comparison of simulation and experimental results shows that the loop-source TEM is more suitable for shallow refined detection in terms of response quality and construction efficiency.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"9450-9461"},"PeriodicalIF":3.4,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10835096","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993851","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Multi-Scale Channel Distillation Network for Image Compressive Sensing 图像压缩感知的多尺度通道蒸馏网络

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-01-09 DOI: 10.1109/ACCESS.2025.3527756

Tianyu Zhang;Kuntao Ye;Yue Zhang;Rui Lu

{"title":"Multi-Scale Channel Distillation Network for Image Compressive Sensing","authors":"Tianyu Zhang;Kuntao Ye;Yue Zhang;Rui Lu","doi":"10.1109/ACCESS.2025.3527756","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3527756","url":null,"abstract":"Recently, convolutional neural networks (CNNs) have demonstrated striking success in computer vision tasks. Methods based on CNNs for image compressive sensing (CS) have also gained prominence. However, existing methods tend to increase the depth of the network in feature space for better reconstruction quality, neglecting the hierarchical representation of intermediate features in pixel space. In order to coordinate the feature space and pixel space to complete the deep reconstruction of images, and further improve the reconstruction performance of current CS methods, we propose a multi-scale channel distillation network (MSCDN). This network first obtains images of multiple scales using a scale-space image decomposition method at the sampling stage, followed by sampling these decomposed images through a convolutional operation. In this way, multi-scale information in the compressed domain is aggregated. During the reconstruction phase, a low-frequency information recovery network generates a preliminary image, whereas a high-frequency feature aggregation network refines the image further. Specifically, we design a dual-branch deep reconstruction architecture with channel distillation residual block (CDRB) as the core component. One branch extracts features gradually by cascading multiple CDRB modules, thereby supplementing the initial reconstructed image with a large amount of high-frequency content in feature space. The other branch takes the initial reconstructed image as input and sequentially fuses the intermediate feature outputs by CDRBs to increase the local details of the image in pixel space. Combining outputs from both branches, we achieve an optimal reconstructed image. Extensive experimental results on four benchmark datasets demonstrate that MSCDN surpasses state-of-the-art CS methods not only in reconstruction accuracy but also in perceptual visual quality.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"9524-9537"},"PeriodicalIF":3.4,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10835084","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993120","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Fusing Emotion and Art Communication: A Disc Interval-Valued Fermatean Fuzzy Decision-Making Approach 情感与艺术沟通的融合：圆盘区间值费马模糊决策方法

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-01-09 DOI: 10.1109/ACCESS.2025.3527580

Liu Peng

引用次数: 0

Projection-Type Holographic Display With Observation Area Expanded Using Time-Division Reproduced Light Generation System 采用分时再现光产生系统扩大观测面积的投影式全息显示器

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-01-09 DOI: 10.1109/ACCESS.2025.3527576

Tomoki Kanno;Yasuyuki Ichihashi;Takashi Kakue;Koki Wakunami;Ryutaro Oi;Tomoyoshi Shimobaba;Tomoyoshi Ito

引用次数: 0

Design of an Ultra Compact 180-Degree Ring Coupler With Harmonic Suppression Using Composite Lines for Modern RF Systems 现代射频系统用复合线谐波抑制超紧凑180度环形耦合器的设计

IF 3.4 3区计算机科学

IEEE Access Pub Date : 2025-01-09 DOI: 10.1109/ACCESS.2025.3527743

Saeed Roshani;Farid Zubir;Mohsen Karimi;Muhammad Akmal Chaudhary;Maher Assaad;Fawwaz Hazzazi;Noorlindawaty Md Jizat;Sobhan Roshani

{"title":"Design of an Ultra Compact 180-Degree Ring Coupler With Harmonic Suppression Using Composite Lines for Modern RF Systems","authors":"Saeed Roshani;Farid Zubir;Mohsen Karimi;Muhammad Akmal Chaudhary;Maher Assaad;Fawwaz Hazzazi;Noorlindawaty Md Jizat;Sobhan Roshani","doi":"10.1109/ACCESS.2025.3527743","DOIUrl":"https://doi.org/10.1109/ACCESS.2025.3527743","url":null,"abstract":"This paper presents lumped reactive capacitors and inductors to create a compact filtering hybrid ring coupler (HRC) with a broad suppression band. The traditional 180-degree ring coupler has six long <inline-formula> <tex-math>$lambda $ </tex-math></inline-formula>/4 branches. In the proposed coupler, six compact proposed branches are applied instead of these six long <inline-formula> <tex-math>$lambda $ </tex-math></inline-formula>/4 branches, resulting in size reduction and harmonic suppression. The designed HRC works at 800 MHz and provide 220 MHz operating bandwidth from 690 MHz from 910MHz, which is equal to 27.5% FBW. The designed HRC has acceptable performance in operating frequency and higher frequencies. The designed 180-degree ring hybrid coupler (HRC) achieves a broad stop-band ranging from 1.7 GHz to 5 GHz, effectively rejecting <inline-formula> <tex-math>$2^{mathrm {nd}}$ </tex-math></inline-formula> to <inline-formula> <tex-math>$6^{mathrm {th}}$ </tex-math></inline-formula> harmonic. The proposed HRC is designed, simulated and fabricated, which the measured results validate the simulations. In the pass band, the proposed hybrid coupler exhibits less than 0.2 dB insertion loss, over 26 dB return loss, and 49 dB isolation measured at 800 MHz. Moreover, the conventional 800 MHz HRC occupied <inline-formula> <tex-math>$0.5~lambda times 0.25~lambda $ </tex-math></inline-formula> (140.7 mm <inline-formula> <tex-math>$times 70.8$ </tex-math></inline-formula> mm), while size of the proposed HRC is 10.7 mm <inline-formula> <tex-math>$times 6.7$ </tex-math></inline-formula> mm (<inline-formula> <tex-math>$0.038~lambda times 0.024~lambda $ </tex-math></inline-formula>), which provides 99% size reduction.","PeriodicalId":13079,"journal":{"name":"IEEE Access","volume":"13 ","pages":"8214-8224"},"PeriodicalIF":3.4,"publicationDate":"2025-01-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=10835077","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142993115","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0