检索结果-内蒙古大学图书馆

Multi-task learning-based temporal pattern matching network for guitar tablature transcription

Neural Computing and Applications 2025年 1-20页

作者： Kim, Taehyeon Kim, Man-Je Ahn, Chang Wook Artificial Intelligence Graduate School Gwangju Institute of Science and Technology 123 Cheomdangwagi-ro Buk-gu Gwangju61005 Korea Republic of Convergence of AI Chonnam National University 77 Yongbong-ro Buk-gu Gwangju61186 Korea Republic of GIST Institute for Artificial Intelligence Gwangju Institute of Science and Technology 123 Cheomdangwagi-ro Buk-gu Gwangju61005 Korea Republic of

Guitar tablature transcription poses unique challenges in automatic music transcription, as it requires capturing both pitch and string usage on a multi-string instrument with various expressive techniques. While guitar tablature is widely used by guitarists in the music field, neural architecture modeling for this task remains underexplored, particularly in accurately mapping pitches to their respective strings. In this work, we propose a multi-task learning-based temporal pattern-matching network (TPMNet) that effectively captures temporal information from guitar recordings, improving the alignment of predicted results. The key contribution of this work is the advancement of neural network architecture, leading to notable improvements in prediction performance for guitar tablature transcription. Additionally, we explored the optimal pooling layer selection method tailored to different tasks, addressing a long-confusing problem in the field. TPMNet’s efficacy was validated through experiments on the GuitarSet dataset, and its generalizability was confirmed via cross-evaluation with the EGDB dataset. © The Author(s), under exclusive licence to Springer-Verlag London Ltd., part of Springer Nature 2025.

关键词： Multi-task learning

来源：评论

学校读者我要写书评

暂无评论

Robust Real-Time Sampling-Based Motion Planner for Autonomous Vehicles in Narrow Environments

引用

IEEE Transactions on Automation science and Engineering 2025年 22卷 16250-16265页

作者： Minsoo Kim Arthur Esquerre-Pourtère Jaeheung Park Department of Intelligence and Information Graduate School of Convergence Science and Technology Seoul National University Seoul Republic of Korea Department of Intelligence and Information Graduate School of Convergence Science and Technology and ASRI AIIS Seoul National University Seoul Republic of Korea Advanced Institute of Convergence Technology Suwon Republic of Korea

Real-time sampling-based planners increasingly use learned sampling distributions for faster planning in autonomous vehicles. These planners employ a neural network to predict the optimal path and bias some samples toward the path. However, inherent prediction inaccuracies of the network often lead to suboptimal paths, especially in narrow spaces. Learned samples should be used carefully based on accuracy, as inaccurate samples can degrade planning performance. To address this problem, this paper proposes Learned Adaptive Anytime TargetTree-RRT* (LA3T*) algorithm. The proposed planner introduces the adaptive biasing ratio. The approach learns to assess the reliability of the learned distribution using the network’s confidence. This confidence approximates a proper ratio of learned samples used, thereby adaptively maximizing planning performance while considering a level of prediction accuracy. Furthermore, the LA3T* algorithm incorporates the target tree algorithm. The goal pose is replaced with a set (target tree) of pre-defined optimal path segments, reducing computational efforts in narrow regions. Experiments in various driving tasks explore the benefits of each component through ablation studies. The proposed algorithm significantly increases the success rate and reduces the path length in simulated and real-world scenarios compared to other sampling-based methods. Note to Practitioners—This work was motivated by the need to develop a practical real-time motion planning algorithm for autonomous vehicles in narrow environments. Although learning-based sampling methods have been actively explored, their planning performance often degrades due to inaccurate predictions from the learned model, limiting their practicability and robustness. The Learned Adaptive Anytime TargetTree-RRT* (LA3T*) algorithm addresses this problem by adaptively leveraging the learned model based on its reliability and integrating it with the target tree algorithm. Experiments conducted in

关键词： Planning Prediction algorithms Real-time systems Neural networks Heuristic algorithms Approximation algorithms Accuracy Autonomous vehicles Vehicle dynamics Training

来源：评论

学校读者我要写书评

暂无评论

Flash‑Induced High‑Throughput Porous Graphene via Synergistic Photo‑Effects for Electromagnetic Interference Shielding

引用

nano-Micro Letters 2023年第11期15卷 1-18页

作者： Jin Soo Lee Jeong‑Wook Kim Jae Hee Lee Yong Koo Son Young Bin Kim Kyoohee Woo Chanhee Lee Il‑Doo Kim Jae Young Seok Jong Won Yu Jung Hwan Park Keon Jae Lee Department of Materials Science and Engineering Korea Advanced Institute of Science and Technology(KAIST)291 Daehak‑roYuseong‑guDaejeon 34141Republic of Korea School of Electrical Engineering Korea Advanced Institute of Science and Technology(KAIST)291 Daehak‑roYuseong‑guDaejeon 34141Republic of Korea Department of Printed Electronics Nano‑Convergence Manufacturing Systems Research DivisionKorea Institute of Machinery and Materials(KIMM)156 Gajeongbuk‑RoYuseong‑GuDaejeon 34103Republic of Korea Department of Mechanical System Design Engineering Seoul National University of Science and Technology232 Gongneung‑roNowon‑guSeoul 01811Republic of Korea Department of Mechanical Engineering(Department of Aeronautics Mechanical and Electronic Convergence Engineering)Kumoh National Institute of Technology61 Daehak‑roGumiGyeongbuk 39177Republic of Korea

Porous 2D materials with high conductivity and large surface area have been proposed for potential electromagnetic interference(EMI)shielding materials in future mobility and wearable applications to prevent signal noise,transmission inaccuracy,system malfunction,and health ***,we report on the synthesis of lightweight and flexible flash-induced porous graphene(FPG)with excellent EMI shielding *** broad spectrum of pulsed flashlight induces photo-chemical and photo-thermal reactions in polyimide films,forming 5×10 cm^(2)-size porous graphene with a hollow pillar structure in a few *** resulting material demonstrated low density(0.0354 g cm^(−3))and outstanding absolute EMI shielding effectiveness of 1.12×10^(5) dB cm^(2) g^(−1).The FPG was characterized via thorough material analyses,and its mechanical durability and flexibility were confirmed by a bending cycle ***,the FPG was utilized in drone and wearable applications,showing effective EMI shielding performance for internal/external EMI in a drone radar system and reducing the specific absorption rate in the human body.

关键词： Porous graphene Flash lamp Photo-pyrolysis High-throughput Electromagnetic interference shielding

来源：评论

学校读者我要写书评

暂无评论

Surface Modification of 3D Printed Biomedical Scaffolds Through an Electron Beam Processe 17

Surface Modification of 3D Printed Biomedical Scaffolds Thro...

引用

17th IEEE International Conference on nano/Molecular Medicine and Engineering, nanoMED 2024

作者： Cha, Goeun Woo, Jonghyeon Noh, Jeongsik Park, Jongsung Kyungpook National University School of Advanced Science and Technology Convergence Graduate School Sangju37224 Korea Republic of Kyungpook National University Department of Precision Mechanical Engineering Sangju37224 Korea Republic of

ISBN: (纸本)9798331516994

This study introduces a method for employing electron beam processing to effectuate both physical and chemical modifications to the surface of 3D printed biomedical scaffolds. Generally, 3D printed scaffolds exhibit certain limitations, including rough surface textures and dimensional inaccuracies, which hinder their applicability in medical contexts. To mitigate these issues, we implemented the electron beam technique to enhance surface smoothness and facilitate chemical alterations. The proposed method is not only efficient in polishing the surfaces of the printed scaffolds but also capable of processing a substantial number of samples within a brief timeframe. This technique holds potential for various biomedical implant devices and can be adapted for use with a range of metallic materials and polymer printing. Notably, the electron beam process described herein relies solely on heat treatment via electron application, thereby ensuring the complete elimination of any potential adverse effects that may arise during clinical trials related to the materials used. © 2024 IEEE.

关键词： Scaffolds (biology)

来源：评论

学校读者我要写书评

暂无评论

Direct Anodic Electrografting of Osmium Complexes on Indium Tin Oxide

SSRN

引用

SSRN 2024年

作者： Lee, Haeyeon Go, Su Yong Kim, Chae Yun Kim, Eun Joong Kang, Chung Mu Chung, Taek Dong Department of Chemistry Seoul National University Seoul08826 Korea Republic of Program in Nano Science and Technology Graduate School of Convergence Science and Technology Seoul National University Suwon16229 Korea Republic of Advanced Institute of Convergence Technology Suwon16229 Korea Republic of

Indium tin oxide (ITO) is a widely used electrode material in electrochemical biosensors owing to its beneficial characteristics such as transparency and chemical inertness. The exceptionally sluggish kinetics of heterogeneous electron transfer in ITO provides an exceptionally low background current, making electron mediators essential for electrochemical sensors. However, the chemical modification of ITO with molecular mediators is still challenging. Few straightforward and credible methods are available to obtain sufficiently thin mediator films firmly anchored on the ITO without losing its inherent benefits, including low background current and transparency. This study demonstrated the anodic electrografting of three osmium complexes, which allowed the formation of molecular mediator monolayers on ITO. The densely packed monolayer of osmium complexes retained their reversible redox processes over hundreds of potential cycles. We confirmed electron mediation by the modified electrodes for aminophenol oxidation, suggesting a simple protocol for developing ITO-based biosensors functionalized with molecular mediators. © 2024, The Authors. All rights reserved.

关键词： Tin oxides

来源：评论

学校读者我要写书评

暂无评论

Facile Surface Functionalization of Electrospun Elastic nanofibers Via Initiated Chemical Vapor Deposition for Enhanced Neural Cell Adhesion and Alignment

引用

Advanced Fiber Materials 2024年第5期6卷 1583-1595页

作者： Yerim Jang Soonjong Roh Younghak Cho Youngmee Jung Kangwon Lee Nakwon Choi Jin Yoo Hyejeong Seong Brain Science Institute Korea Institutes of Science and Technology(KIST)Seoul 02792Republic of Korea KU-KIST Graduate School of Converging Science and Technology Korea UniversitySeoul 02841Republic of Korea Biomaterials Research Center Korea Institutes of Science and Technology(KIST)Seoul 02792Republic of Korea Department of Applied Bioengineering Graduate School of Convergence Science and TechnologySeoul National UniversitySeoul 08826Republic of Korea Research Institute for Convergence Science Seoul National UniversitySuwon-siGyeonggi-do 16229Republic of Korea Division of Bio-Medical Science and Technology KIST SchoolKorea University of Science and Technology(UST)Seoul 02792Republic of Korea School of Electrical Electronic Engineering YU-KIST InstituteYonsei UniversitySeoul 03722Republic of Korea

An advanced approach for functionalizing the surfaces of electrospun poly(l-lactide-co-ε-caprolactone)(PLCL)nanofibers for biomedical applications is presented *** initiated chemical vapor deposition(iCVD),a coating of the copolymer p(PFMA-co-DVB)containing poly(pentafluorophenyl methacrylate)(PFMA)and divinylbenzene(DVB)was applied to the PLCL *** coating facilitated efficient immobilization of the biomolecules on the PLCL nanofiber surfaces,allowing precise adjustments to the polymer composition through modulation of the monomer flow *** resulting copolymer exhibited superior efficiency for immobilizing IgG,as confirmed by immunofluorescence intensity *** vitro studies conducted with different neural cell types demonstrated that the laminin-coated iCVD-functionalized PLCL nanofibers maintained their inherent biocompatibility while significantly enhancing cell *** exploiting the elastic nature of the PLCL nanofibers,cell elongation could be successfully manipulated by controlling the nanofiber alignment,as demonstrated by scanning electron microscopy and quantification of the immunofluorescence image *** findings highlight the potential of iCVD-modified PLCL nanofibers as versatile platforms for neural tissue engineering and various biomedical applications,allowing valuable biomaterial surface modifications for enhanced cellular interactions.

关键词： Electrospun nanofiber Initiated chemical vapor deposition Poly(l-lactide-co-ε-caprolactone) Post-surface modification Neural cells

来源：评论

学校读者我要写书评

暂无评论

Polymerization of pyrrole induced by pillararene functionalized graphene for supercapacitor electrode

引用

Chinese Chemical Letters 2022年第11期33卷 4846-4849页

作者： Fang Guo Junqiang Guo Zhiqiang Zheng Tao Xia Aadil Nabi Chishti Liwei Lin Wang Zhang Guowang Diao School of Chemistry and Chemical Engineering Yangzhou UniversityYangzhou 225002China Department of Applied Bioengineering Graduate School of Convergence Science and TechnologySeoul National UniversitySuwon 443-270South Korea

Conducting polymer is an important electrode material for supercapacitors because of its high initial specific ***,a novel nanocomposite composed of polypyrrole(PPy)film homogeneously immobilized on the pillar[5]arene functionalized reduced graphene oxide nanosheets(RGO-HP5 A-PPy)was successfully ***-HP5 A induced pyrrole to polymerize on the graphene surface and the specific capacitance loss caused by PPy agglomeration was ***,the specific capacitance of RGOHP5 A-PPy was up to 495 F/g at 1 A/*** with pure PPy(319 F/g),the specific capacitance was increased by 55%.The specific capacitance retention of the assembled symmetric supercapacitor reached76%after 10,000 cycles at 5 A/*** study gave full play to the advantages of pillar[5]arene,graphene and PPy,and was expected to promote the development of supramolecular functionalized composites in energy storage.

关键词： nanocomposite Pillar[5]arene Graphene Polypyrrole Supercapacitor

来源：评论

学校读者我要写书评

暂无评论

Thermal reaction characterization of nano-aluminum powder at different heating rates by synchronous thermal analysis

引用

Journal of Thermal Analysis and Calorimetry 2023年第11期148卷 4937-4947页

作者： Qu, Jiao Deng, Jun Luo, Zhen-Min Xiao, Yang Shu, Chi-Min School of Safety Science and Engineering Xi’an University of Science and Technology No. 58 Yanta Mid. Rd. Shaanxi Xi’an710054 China Postdoctoral Program Xi’an University of Science and Technology No. 58Yanta Mid. Rd. Shaanxi Xi’an710054 China Shaanxi Key Laboratory of Prevention and Control of Coal Fire No. 58 Yanta Mid. Rd. Shaanxi Xi’an710054 China Graduate School of Engineering Science and Technology National Yunlin University of Science & Technology 123 University Rd. Sec. 3 Yunlin Douliou64002 Taiwan

This article reports an investigation on the thermal reactivity of nano-aluminum powder (NAP) in the air. We performed tests on the metallic powder in synchronous thermal analysis at four heating rates of 5, 10, 15, and 20 °C min−1. Given the polymorphic phase change of aluminum, the oxidation process of NAP was divided into four stages. The results showed a notable effect from the heating rate on the oxidation of NAP. The mass gain decreased at stage II and increased at stage IV as the rate of heating increased. The maximum mass gain rate at 5, 10, 15, and 20 °C min−1 was 2.38, 4.70, 33.79, and 53.38% min−1, respectively, while the thermal release was 35,126, 27,955, 15,608, and 16,336 J g−1, respectively. Furthermore, the average apparent activation energy, preexponential factor, and mechanism function were obtained by the kinetic integral method. © 2023, Akadémiai Kiadó, Budapest, Hungary.

关键词： Heating rate

来源：评论

学校读者我要写书评

暂无评论

Privacy-Enhanced Data Deduplication Computational Intelligence Technique for Secure Healthcare Applications

引用

Computers, Materials & Continua 2022年第2期70卷 4169-4184页

作者： Jinsu Kim Sungwook Ryu Namje Park Department of Convergence Information Security Graduate SchoolJeju National University63294Korea Master’s Program in Future Strategy Korea Advanced Institute of Science and Technology34141Korea Department of Computer Education Teachers CollegeJeju National University63294Korea

A significant number of cloud storage environments are already implementing deduplication *** to the nature of the cloud environment,a storage server capable of accommodating large-capacity storage is *** storage capacity increases,additional storage solutions are *** leveraging deduplication,you can fundamentally solve the cost ***,deduplication poses privacy concerns due to the structure *** this paper,we point out the privacy infringement problemand propose a new deduplication technique to solve *** the proposed technique,since the user’s map structure and files are not stored on the server,the file uploader list cannot be obtained through the server’s meta-information analysis,so the user’s privacy is *** addition,the personal identification number(PIN)can be used to solve the file ownership problemand provides advantages such as safety against insider breaches and sniffing *** proposed mechanism required an additional time of approximately 100 ms to add a IDRef to distinguish user-file during typical deduplication,and for smaller file sizes,the time required for additional operations is similar to the operation time,but relatively less time as the file’s capacity grows.

关键词： Computational intelligence cloud multimedia data deduplication

来源：评论

学校读者我要写书评

暂无评论

Deep learning-based virtual staining,segmentation,and classification in label-free photoacoustic histology of human specimens

引用

Light(science & Applications) 2024年第10期13卷 2353-2366页

作者： Chiho Yoon Eunwoo Park Sampa Misra Jin Young Kim Jin Woo Baik Kwang Gi Kim Chan Kwon Jung Chulhong Kim Departments of Electrical Engineering Convergence IT EngineeringMechanical EngineeringMedical Science and EngineeringGraduate School of Artificial Intelligenceand Medical Device Innovation CenterPohang University of Science and Technology(POSTECH)PohangRepublic of Korea Opticho Inc. PohangRepublic of Korea Department of Health Sciences and Technology Gachon Advanced Institute for Health Sciences and Technology(GAIHST)Gachon UniversityIncheonRepublic of Korea Cancer Research Institute College of MedicineThe Catholic University of KoreaSeoulRepublic of Korea Department of Hospital Pathology Seoul St.Mary’s HospitalCollege of MedicineThe Catholic University of KoreaSeoulRepublic of Korea

In pathological diagnostics,histological images highlight the oncological features of excised specimens,but they require laborious and costly staining *** recent innovations in label-free microscopy that simplify complex staining procedures,technical limitations and inadequate histological visualization are still problems in clinical ***,we demonstrate an interconnected deep learning(DL)-based framework for performing automated virtual staining,segmentation,and classification in label-free photoacoustic histology(PAH)of human *** framework comprises three components:(1)an explainable contrastive unpaired translation(E-CUT)method for virtual H&E(VHE)staining,(2)an U-net architecture for feature segmentation,and(3)a DL-based stepwise feature fusion method(StepFF)for *** framework demonstrates promising performance at each step of its application to human liver *** virtual staining,the E-CUT preserves the morphological aspects of the cell nucleus and cytoplasm,making VHE images highly similar to real H&E *** segmentation,various features(e.g.,the cell area,number of cells,and the distance between cell nuclei)have been successfully segmented in VHE ***,by using deep feature vectors from PAH,VHE,and segmented images,StepFF has achieved a 98.00%classification accuracy,compared to the 94.80%accuracy of conventional PAH *** particular,StepFF’s classification reached a sensitivity of 100%based on the evaluation of three pathologists,demonstrating its applicability in real clinical *** series of DL methods for label-free PAH has great potential as a practical clinical strategy for digital pathology.

关键词： stepwise Deep spite

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：