연구성과로 돌아가기
2023 연구성과 (244 / 285)
※ 컨트롤 + 클릭으로 열별 다중 정렬 가능합니다.
Excel 다운로드
| WoS | SCOPUS | Document Type | Document Title | Abstract | Authors | Affiliation | ResearcherID (WoS) | AuthorsID (SCOPUS) | Author Email(s) | Journal Name | JCR Abbreviation | ISSN | eISSN | Volume | Issue | WoS Edition | WoS Category | JCR Year | IF | JCR (%) | FWCI | FWCI Update Date | WoS Citation | SCOPUS Citation | Keywords (WoS) | KeywordsPlus (WoS) | Keywords (SCOPUS) | KeywordsPlus (SCOPUS) | Language | Publication Stage | Publication Year | Publication Date | DOI | JCR Link | DOI Link | WOS Link | SCOPUS Link |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ○ | Conference paper | An Analysis of Research Trends on Language Model Using BERTopic | Although language models have played a crucial role in various natural language processing tasks, there has been little research that focuses on systematic analysis and review of research topic trends in these models. In this paper, we conducted a comprehensive analysis of 31 years of research trends in the field of language models, using publications from Scopus, an internationally renowned academic database, to identify research topics related to language models. We adopted BERTopic, a state-of-the-art topic modeling technique, on the 13,754 research articles about language models. The research on language models has gradually increased since 1991, and there is a sudden increase in the number of publications with the emergence of BERT and GPT in 2018. We assigned 14 main topics with meaningful keywords clustered by BERTopic model. Among 14 topics, research related to speech recognition, statistical language models, and pre-trained language models demonstrated the most vigorous research fields. Our results demonstrate a more systematic and comprehensive trend in language model research, which is expected to provide an important foundation for future research directions. © 2023 IEEE. | Kang, Woojin; Kim, Yumi; Kim, Heesop; Lee, Jongwook | Kyungpook National University, Department of Library and Information Science, Daegu, South Korea; Kyungpook National University, Department of Library and Information Science, Daegu, South Korea; Kyungpook National University, Department of Library and Information Science, Daegu, South Korea; Kyungpook National University, Department of Library and Information Science, Daegu, South Korea | 58999140800; 59052191800; 8865330100; 54988756200 | Proceedings - 2023 Congress in Computer Science, Computer Engineering, and Applied Computing, CSCE 2023 | 1.2 | 2025-06-25 | 2 | BERT; language models; re-search trends; Short Research Paper; topic modeling | Computational linguistics; Modeling languages; Natural language processing systems; BERT; Language model; Language processing; Natural languages; Re-search trend; Research papers; Research topics; Research trends; Short research paper; Topic Modeling; Speech recognition | English | Final | 2023 | 10.1109/csce60160.2023.00032 | 바로가기 | 바로가기 | |||||||||||||||||||
| ○ | ○ | Proceedings Paper | An Effective IoT based Waste Management System using Fuzzy Soft Expert Systems | The rapid population growth in urban areas leads to increased waste generation and the need for sustainable waste management solutions to preserve the global environment. In this paper, we present an intelligent garbage collection system for smart city environments, which consists of three key components. First, we propose a novel model to optimize the positioning of garbage bins to prevent both overflow and underutilization. We leverage a fuzzy soft expert set-based solution to identify optimal locations within a selected region. Experimental results demonstrate that our proposed method significantly outperforms manual placement strategies in mitigating issues arising from inappropriate bin placement. Second, we address the problem of improper waste collection from overflowing garbage bins by developing a smart bin system. This system employs a voice-controlled bin that sends automatic alerts to collectors before reaching its full capacity, as well as audio alerts for the public to prevent waste overflow around the bin. Lastly, we integrate a mobile application and cloud-based database to enable real-time monitoring of bin statuses. This comprehensive approach to waste management in urban areas not only enhances the efficiency of garbage collection but also contributes to a cleaner and more sustainable environment. | Lavanya, G.; Ganeshkumar, P.; Vignesh, S.; Paul, Anand | Saveetha Engn Coll, Dept Artificial Intelligence & Data Sci, Chennai, Tamil Nadu, India; Coll Engn, Dept Comp Sci & Engn, Chennai, Tamil Nadu, India; Saveetha Engn Coll, Dept Mech Engn, Chennai, Tamil Nadu, India; Kyungpook Natl Univ, Sch Comp Sci & Engn, Seoul, South Korea | R, Raja/HTS-3965-2023; Paul, Anand/V-6724-2017; G, Lavanya/AAN-7839-2021 | 19933862200; 57203113010; 57211206969; 56650522400 | glavanya@saveetha.ac.in;ganesh23508@gmail.com;vigneshs@saveetha.ac.in;anand@knu.ac.kr; | 2023 IEEE INTERNATIONAL CONFERENCE ON FUZZY SYSTEMS, FUZZ | 1544-5615 | 0 | 2025-06-25 | 0 | 0 | Smart city; Internet of Things; Cloud database; Fuzzy soft expert set | COLLECTION | Cloud database; Fuzzy soft expert set; Internet of Things; Smart city | Bins; Environmental management; Expert systems; Internet of things; Population statistics; Sustainable development; Urban growth; Waste management; Cloud database; Collection systems; Fuzzy soft expert set; Garbage collection; Global environment; Rapid population growth; Sustainable waste management; Urban areas; Waste generation; Waste management systems; Smart city | English | 2023 | 2023 | 10.1109/fuzz52849.2023.10309689 | 바로가기 | 바로가기 | 바로가기 | |||||||||||
| ○ | Conference paper | An efficient network operation automation scheme using network status information for local 5G networks | 5G SA communication technology is being recognized as a crucial technology shaping the future of wireless communication. Compared to previous generations of networks, 5G offers ultra-high-speed, low latency, and massive connectivity, presenting both enterprises and individual users with opportunities for new business models and an enhanced quality of life. One of the most significant changes between the conventional mobile communication environment and 5G is the activation of various application cases in B2B services based on the local 5G network. The local 5G network, tailored to specific sites, requires effective management of high-volume data generated in dense areas and necessitates minimizing manual intervention by administrators. The ZSM (Zero-touch Service Management) technology is gaining attention to meet these requirements. ZSM focuses on automated operations for managing large systems like commercial mobile communication networks. It is anticipated that ZSM will also be applied to the local 5G network for efficient management and operation of tailored services. This paper proposes a network management technique that minimizes administrator intervention in the local 5G network environment. The proposed network management technique collects statistical logs from local 5G network base stations and core systems, supports real-time network status checks as well as traffic predictions using machine learning. Additionally, it offers automated operation techniques for the local 5G network by incorporating functions such as disconnecting network access for devices that are factors of malfunction within the specialized network. © 2023 IEEE. | Yoon, Mahnsuk; Seo, Jihun; Cho, Kyucheol | GERI(Gumi Electronics & Information Technology Research Institute), Future Mobile Communication Research Center, Gumi, South Korea; Korea Transport Institute, Center for Connected & Automated Driving Research, Department of the Fourth Industrial Revolution & Transport, Sejong, South Korea; Kyungpook National University, School of Computer Secience and Engineering, Software Education Center, Daegu, South Korea | 57203640624; 58000906600; 26031217700 | yms@geri.re.kr; | International Conference on ICT Convergence | 2162-1233 | 0.43 | 2025-06-25 | 1 | Automation of operations; Local 5G Networks; Network Management; Zero-touch Service Management | Automation; Information management; Network management; Queueing networks; Real time systems; Automated operations; Automation of operation; Automation scheme; Local 5g network; Network management techniques; Network operations; Network status; Networks management; Service management; Zero-touch service management; 5G mobile communication systems | English | Final | 2023 | 10.1109/ictc58733.2023.10392675 | 바로가기 | 바로가기 | |||||||||||||||||
| ○ | ○ | Proceedings Paper | An End-to-End Motion Planner Using Sensor Fusion for Autonomous Driving | Autonomous driving vehicles and advanced driver-assistance systems are gaining tremendous attention with the hope of providing a new transportation mode that is more convenient and ensures road safety. Different types of sensors are deployed together with the aid of deep learning (DL) techniques to help the vehicle perceive the surrounding environment and navigate toward the destination. In this paper, we implemented a deep learning-based motion planner using sensor fusion from LiDAR point clouds and camera RGB images to predict future waypoints. The model is trained in an end-to-end manner in which input are the multimodal sensor data, and output is the predicted future waypoints. A transformer module with a self-attention mechanism is used to integrate the representation of the two sensor modalities. During training, auxiliary tasks including depth estimation and bird-eye-view semantic segmentation are carried out to provide an intermediate representation of the perception process as well as to enhance the performance of the motion planning task. Experimental results obtained from different model configurations on the Longest6 benchmark have shown that our proposed model achieves competitive performance compared to baselines. | Thu, Nguyen Thi Hoai; Han, Dong Seog | Kyungpook Natl Univ, Grad Sch, Sch Elect & Elect Engn, Daegu, South Korea; Kyungpook Natl Univ, Sch Elect Engn, Daegu, South Korea | Nguyen, Thu/AAC-1112-2021 | 57221708543; 7403219442 | thunguyen@knu.ac.kr;dshan@knu.ac.kr; | 2023 INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE IN INFORMATION AND COMMUNICATION, ICAIIC | 2831-6991 | 1.94 | 2025-06-25 | 1 | 4 | Autonomous vehicles; Motion planning; Sensor fusion; End-to-end deep learning | Autonomous vehicles; End-to-end deep learning; Motion planning; Sensor fusion | Advanced driver assistance systems; Automobile drivers; Benchmarking; Deep learning; Motion planning; Motor transportation; Semantic Segmentation; Semantics; Autonomous driving; Autonomous Vehicles; End to end; End-to-end deep learning; Motion planners; Motion-planning; Road safety; Sensor fusion; Transportation mode; Waypoints; Autonomous vehicles | English | 2023 | 2023 | 10.1109/icaiic57133.2023.10067069 | 바로가기 | 바로가기 | 바로가기 | ||||||||||||
| ○ | Conference paper | An Enhanced WIFI Indoor Positioning Method Based on SNGAN | With the development of wireless communication technology research, location-based services (LBS) have emerged due to provide useful information to users based on the current user locations or a given location-indoor positioning technology has become a hot spot for research. Indoor positioning technology based on WIFI signals attentions owing to WIFI networks cover various indoor environments. Received signal strength (RSS) based indoor positioning technology has become an effective indoor positioning technology as a method that utilizes the wireless signal propagation model. However, due to the complexity and variability of the indoor environment, RSS values are highly vulnerable to noise and multipath interference, RSS-based WIFI fingerprinting can effectively avoid this problem. During the construction of RSS fingerprint maps, fingerprint data collection is time-consuming and laborious, how to use a small amount of fingerprint data to achieve high-precision positioning becomes a difficult point for fingerprint positioning technology. To solve this problem, a high-precision indoor localization method based on spectral normalization for generative adversarial networks (SNGAN) is proposed. The generative adversarial network with spectral normalization not only retains the original information to the greatest extent during the training process, but also makes the generated data closer to the distribution of real data and speeds up the training. Collecting a small amount of fingerprint data achieves higher localization accuracy. © 2023 Proceedings of the 36th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS+ 2023. All rights reserved. | Cui, Shuyu; Dong, Jiabin; Hwang, Jun Gyu; Rana, Lila; Li, JinLong; Park, Joon Goo | Kyungpook National University, South Korea; Kyungpook National University, South Korea; Kyungpook National University, South Korea; Kyungpook National University, South Korea; Kyungpook National University, South Korea; Kyungpook National University, South Korea | 58783970500; 58784428300; 55816210200; 58783364500; 58783810700; 24329712800 | Proceedings of the 36th International Technical Meeting of the Satellite Division of the Institute of Navigation, ION GNSS+ 2023 | 0.46 | 2025-06-25 | 1 | deep learning; fingerprint localization; generative adversarial networks (GAN); synthetic data; wireless sensor networks | Deep learning; Generative adversarial networks; Global positioning system; Indoor positioning systems; Location; Navigation; Satellites; Telecommunication services; Wi-Fi; Wireless local area networks (WLAN); Wireless sensor networks; Deep learning; Fingerprint localization; Generative adversarial network; Indoor environment; Indoor positioning; Localisation; Positioning technologies; Received signal strength; Spectral normalization; Synthetic data; Location based services | English | Final | 2023 | 10.33012/2023.19217 | 바로가기 | 바로가기 | |||||||||||||||||||
| ○ | Article | An Exploration of Tennis Consumption Experiences by a Beginner Named 'Te-rin-e': The Desire for Self-Expression, Conflict with Mainstream Culture, and Knowledge-Based Consumption | PURPOSE This study focuses on the experiences of novice tennis enthusiasts in Korea, aligning with the rising popularity of tennis as a burgeoning cultural phenomenon. METHODS Employing an ethnographic approach, we included eight participants with <5 years of tennis, in addition to three auxiliary participants. RESULTS The primary findings of this study are as follows: First, novice tennis players from the MZ generation actively employ tennis as a means of personal expression, sharing their tennis-related fashion and experiences on SNS. Second, most novice tennis players report a sense of detachment from the traditional club-centric tennis culture, gravitating toward casual court usage, spontaneous activities, and flexible membership structures. F inally, despite their modest technical proficiency, they avidly engage in knowledge-based consumption, displaying a profound grasp of tennis-related information. CONCLUSIONS These findings provide valuable insights into comprehending this burgeoning cultural phenomenon in Korean society, stemming from the escalating interest in tennis, and can serve as a valuable benchmark for establishing development strategies for tennis. © Korea Institute of Sport Science. | Lim, Seami; Jeong, Howon; Kim, Hanbeom | Korea Institute of Sports Science, South Korea; Kyungpook National University, South Korea; Hankyung National University, South Korea | 57218093074; 58667288900; 57679901400 | hwj2012@knu.ac.kr; | Korean Journal of Sport Science | 1598-2920 | 34 | 3 | 0 | 2025-06-25 | 0 | Conflict; Knowledge-based consumption; Self-expression; Ta-rin-e; Tennis | Korean | Final | 2023 | 10.24985/kjss.2023.34.3.456 | 바로가기 | 바로가기 | ||||||||||||||||
| ○ | Proceedings Paper | An extensive analysis of the sub-parsec region of 3C 84 | The study of jet launching in AGN is an important research method to better understand super-massive black holes (SMBHs) and their immediate surroundings. The main theoretical jet launching scenarios invoke either magnetic field lines anchored to the black holes's (BH) accretion disc [1] or a magnetic field, which is directly connected to its rotating ergosphere [2]. The nearby and bright radio galaxy 3C 84 (NGC1275) is a very suitable target for testing different jet launching mechanisms, aswell as for the study of the innermost, sub-parsec scale AGN structure and the jet origin. Very long baseline interferometry (VLBI) - specifically at millimetre wavelengths - offers an unparalleled view into the physical processes in action, in the close vicinity of SMBHs. Utilising such mm-VLBI observations of 3C 84, we study the jet kinematics of the VLBI core region of 3C 84 by employing all available, high sensitivity 3 mm-VLBI data sets of this source. As part of this analysis we associate the component ejection events with the variability light-curves at different radio frequencies and in the.. -rays. Furthermore, by cross-correlating these light-curves, we determine their time-lags and draw conclusions regarding the location of the high energy emission close to the jet base. | Paraschos, George F.; Kim, J-Y; Krichbaum, T. P.; Oh, J.; Hodgson, J. A.; Gurwell, M. A.; Zensus, J. A. | Max Planck Inst Radioastron, Hugel 69, D-53121 Bonn, Germany; Korea Astron & Space Sci Inst, 776 Daedeokdae Ro, Daejeon 30455, South Korea; Sejong Univ, Dept Phys & Astron, 209 Neungdong Ro, Seoul 05006, South Korea; Harvard & Smithsonian, Ctr Astrophys, 60 Garden St, Cambridge, MA 02138 USA; Kyungpook Natl Univ, Dept Astron & Atmospher Sci, Daegu 702701, South Korea; Joint Inst VLBI ERIC, Oude Hoogeveensedijk 4, NL-7991 PD Dwingeloo, Netherlands | Hodgson, Jeffrey/AAU-9026-2021; Kim, Jae-Young/IUO-6466-2023 | gfparaschos@mpifr-bonn.mpg.de; | 15TH EUROPEAN VLBI NETWORK MINI-SYMPOSIUM AND USERS' MEETING | 0 | GAMMA-RAY EMISSION; ACTIVE GALACTIC NUCLEI; RELATIVISTIC JETS; RADIO; VARIABILITY; KINEMATICS; BLAZARS | English | 2023 | 2023 | 바로가기 | |||||||||||||||||||||||
| ○ | Conference paper | An extensive analysis of the sub-parsec region of 3C 84 | The study of jet launching in AGN is an important research method to better understand supermassive black holes (SMBHs) and their immediate surroundings. The main theoretical jet launching scenarios invoke either magnetic field lines anchored to the black holes's (BH) accretion disc [1] or a magnetic field, which is directly connected to its rotating ergosphere [2]. The nearby and bright radio galaxy 3C 84 (NGC1275) is a very suitable target for testing different jet launching mechanisms, aswell as for the study of the innermost, sub-parsec scaleAGNstructure and the jet origin. Very long baseline interferometry (VLBI)-specifically at millimetre wavelengths-offers an unparalleled view into the physical processes in action, in the close vicinity of SMBHs. Utilising such mm-VLBI observations of 3C 84, we study the jet kinematics of the VLBI core region of 3C 84 by employing all available, high sensitivity 3 mm-VLBI data sets of this source. As part of this analysis we associate the component ejection events with the variability light-curves at different radio frequencies and in the γ-rays. Furthermore, by cross-correlating these light-curves, we determine their time-lags and draw conclusions regarding the location of the high energy emission close to the jet base. © 2023 Sissa Medialab Srl. All rights reserved. | Paraschos, G.F.; Kim, J.-Y.; Krichbaum, T.P.; Oh, J.; Hodgson, J.A.; Gurwell, M.A.; Zensus, J.A. | Max-Planck-Institut för Radioastronomie, Auf dem Högel 69, Bonn, Bonn, D-53121, Germany; Max-Planck-Institut för Radioastronomie, Auf dem Högel 69, Bonn, Bonn, D-53121, Germany, Department of Astronomy and Atmospheric Sciences, Kyungpook National University, Daegu, 702-701, South Korea; Max-Planck-Institut för Radioastronomie, Auf dem Högel 69, Bonn, Bonn, D-53121, Germany; Department of Physics and Astronomy, Sejong University, 209 Neungdong-ro, Gwangjin-gu, Seoul, 05006, South Korea, Joint Institute for VLBI ERIC, Oude Hoogeveensedijk 4, Dwingeloo, 7991 PD, Netherlands; Korea Astronomy and Space Science Institute, 776 Daedeokdae-ro, Yuseong-gu, Daejeon, 30455, South Korea, Department of Physics and Astronomy, Sejong University, 209 Neungdong-ro, Gwangjin-gu, Seoul, 05006, South Korea; Center for Astrophysics | Harvard & Smithsonian, 60 Garden Street, Cambridge, 02138, MA, United States; Max-Planck-Institut för Radioastronomie, Auf dem Högel 69, Bonn, Bonn, D-53121, Germany | 57224966143; 57211836467; 7004015011; 56318374100; 55622269600; 6602556944; 7004929291 | Proceedings of Science | 1824-8039 | 428 | 0 | 2025-06-25 | 0 | Black holes; Galaxies; Gamma rays; Gravitation; Launching; Millimeter waves; Radio astronomy; Stars; Black-hole accretion discs; Ergosphere; Light curves; Magnetic field line; Magnetic-field; Millimeter wavelength; Radio galaxies; Research method; Supermassive black holes; Very long baseline interferometry; Magnetic fields | English | Final | 2023 | 바로가기 | ||||||||||||||||||||
| ○ | Conference paper | An I/O Simulation Method for AUTOSAR-Based Operation Verification in an QEMU-Based Virtual ECU | As the user requirements for automotive embedded systems diversify, hardware performance improves, and software complexity increases. It is, therefore, imperative to find ways to improve safety and reliability of vehicles. Due to these challenges, research has been actively conducted in recent years on methods of verifying automotive embedded systems in a virtual environment, which is highly efficient in terms of time and cost. This paper proposes a method for verifying automotive embedded systems by virtualizing electronic control units (ECUs) through Quick EMUlation (QEMU). The proposed method simulates inputs and outputs of the I/O module by modifying or reading the values of the virtualized ECU's registers. For this purpose, separate Complex Device Driver (CDD) and Application Software (ASW) are configured in AUTomotive Open System Architecture (AUTOSAR). CDD is implemented to periodically input signals corresponding to "on"and "off"to the registers associated with the button inputs of the virtualized ECU. ASW is implemented to periodically read the registers corresponding to the buttons and turn on/off the LED. As a result, we have found that the LED is turned on/off appropriately in the ASW according to the register operation of the virtual ECU in the CDD. This allows software and hardware operation verification through hardware input/output solely by manipulating ECU registers, without the need for a separate external device implementation. Therefore, as a cost-efficient and feasible verification technique, the proposed method can be utilized for future operational verification of vehicle embedded systems based on various input/output scenarios. © 2023 IEEE. | Yang, Anna; Han, Woo Jin; Kim, Hyeong Rae; Kim, Jae Gon; Kum, Daehyun; Cho, Jeonghun | DRIMAES, Korea Aerospace Univ, Sdv Center, Seoul, South Korea; Drimaes, Sdv Center, Seoul, South Korea; Kyungpook National University, School of Electronics Engineering, Daegu, South Korea; Korea Aerospace Univ, School of Electronics and Information Engineering, Goyang, South Korea; Dgist, Division of Automotive Technology, Daegu, South Korea; Kyungpook National Univ, School of Electronics and Engineering, Daegu, South Korea | 59065812900; 58876200300; 57203064517; 56104408600; 16643117800; 7403536198 | jgkim@kau.ac.kr;kumdh@dgist.ac.kr;jcho@knu.ac.kr; | International Conference on ICT Convergence | 2162-1233 | 0.43 | 2025-06-25 | 1 | Automotive Embedded Software; AUTOSAR classic; I/O simulation; QEMU; virtual ECU | Control systems; Embedded systems; Light emitting diodes; Open systems; Verification; Virtual reality; Automotive embedded software; Automotive embedded systems; Automotive open system architecture classic; Automotives; Electronics control unit; I/O simulation; Open system architecture; Quick emulation; Virtual electronic control unit; Application programs | English | Final | 2023 | 10.1109/ictc58733.2023.10393754 | 바로가기 | 바로가기 | |||||||||||||||||
| ○ | Conference paper | An Improved Stacked Multicell Converter with Fault Tolerant Capability Using Finite-Control-Set Model Predictive Control | The stacked multicell (SMC) converter suffers from low-frequency voltage fluctuations in the neutral point (NP), which gets substantially larger at specific operating conditions and affects the safe operation of the converter. This paper proposes an improved SMC (ISMC) converter to cope with this shortcoming. A finite control set model predictive control (FCS-MPC) strategy is designed to accurately track the output current and voltage balancing across the flying capacitors (FC). With this strategy, a fault-tolerant ability of the proposed converter is also presented. A post-fault reconfiguration of the controller is proposed to achieve FC voltage balancing under bidirectional switch faults. Simulation results under steady-state, dynamic, and faulty operating conditions justify the feasibility of the proposed converter. © 2023 IEEE. | Yuan, Zhige; Faraji, Faramarz; Ghias, Amer M. Y. M.; Cha, Honnyong | School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore; School of Energy Engineering, Kyungpook National University, Daegu, South Korea; School of Electrical and Electronic Engineering, Nanyang Technological University, Singapore; School of Energy Engineering, Kyungpook National University, Daegu, South Korea | 58651219800; 57191226987; 46062041900; 24450248400 | 2023 IEEE 3rd International Conference on Industrial Electronics for Sustainable Energy Systems, IESES 2023 | 0.83 | 2025-06-25 | 1 | finite control set model predictive control; multilevel converter; SMC | Fault tolerance; Power converters; Power electronics; Predictive control systems; Finite control set; Finite control set model predictive control; Flying capacitor; Model-predictive control; Multicell; Multilevel converter; Operating condition; Set models; Stacked multicell; Stacked multicell converters; Model predictive control | English | Final | 2023 | 10.1109/ieses53571.2023.10253684 | 바로가기 | 바로가기 | |||||||||||||||||||
| ○ | ○ | Proceedings Paper | An Integrated Approach to Near-duplicate Image Detection | Near-duplicate image detection is a task to find clusters of images that are considered to be the same pictures in human view. This is important in image recommendation systems, because when the systems recommend candidate images, redundancies of retrieved candidate images need to be avoided. In addition, in the era of big-data where image data is overflowing, its importance in terms of saving storage resources further increases. In this paper, we propose a robust model for detecting various types of near-duplicate images by integrating four different detection modules, where we use multiple image feature extractors such as Gabor filter and deep networks. The four modules are then integrated to conduct the multivariate log-likelihood ratio test for detecting duplication. Through computational experiments, we confirmed that our method reaches state-of-the-art performance. | Yang, Heesung; Park, Hyeyoung | Kyungpook Natl Univ, Sch Comp Sci & Engn, Daegu, South Korea | 58175679600; 55713613500 | hs.yang@knu.ac.kr;hypark@knu.ac.kr; | 2023 INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE IN INFORMATION AND COMMUNICATION, ICAIIC | 2831-6991 | 1.45 | 2025-06-25 | 2 | 3 | deep learning features; feature integration; near-duplicate image detection; image recommendation system | deep learning features; feature integration; image recommendation system; near-duplicate image detection | Deep learning; Digital storage; Feature extraction; Gabor filters; Deep learning feature; Feature integration; Image data; Image recommendation system; Image redundancy; Integrated approach; Near-duplicate image detection; Near-duplicates; Robust modeling; Storage resources; Recommender systems | English | 2023 | 2023 | 10.1109/icaiic57133.2023.10067005 | 바로가기 | 바로가기 | 바로가기 | |||||||||||||
| ○ | Conference paper | An Optimal Re-parametrization Scheme for Generalization in Reinforcement Learning | The use of Reinforcement learning (RL) in addressing a variety of engineering tasks has seen a lot of increase lately. However, because RL policies are trained under specific environmental conditions, their performance degrades when deployed in environments with different environmental conditions. As a result, several researchers have focused on developing schemes aimed at minimizing this performance degradation and consequently realizing RL agents that are capable of generalizing to different environmental conditions. In this paper, we propose an optimal re-parameterization scheme based on weighted averaging to facilitate generalization in RL. In the proposed method, two RL agents are trained on simulated environments with different environmental or model parameters, and then covariance matrix adaptation evolution strategies (CMA-ES) is used to determine the optimal averaging weights to combine the two agents for a test environment. We evaluate the performance of our method on a set of popular RL locomotion environments and show that it can significantly improve the generalization performance of RL policies. © 2023 IEEE. | Ivan, Dzeuban Fenyom; Ajani, Oladayo S.; Mallipeddi, Rammohan | Kyungpook National University, Department of Artificial Intelligence, Daegu, South Korea; Kyungpook National University, Department of Artificial Intelligence, Daegu, South Korea; Kyungpook National University, Department of Artificial Intelligence, Daegu, South Korea | 58655933000; 57465126000; 25639919900 | ivanfenyom@gmail.com; | International Conference on ICT Convergence | 2162-1233 | 0 | 2025-06-25 | 0 | Covariance Matrix Adaptation Evolution Strategies; Model Averaging; Reinforcement Learning; Sim-to-real Transfer | Covariance matrix; Evolutionary algorithms; Learning systems; Optimization; Covariance matrix adaptation evolution strategies; Environmental conditions; Generalisation; Learning policy; Model averaging; Parametrizations; Performance; Reinforcement learning agent; Reinforcement learnings; Sim-to-real transfer; Reinforcement learning | English | Final | 2023 | 10.1109/ictc58733.2023.10393351 | 바로가기 | 바로가기 | |||||||||||||||||
| ○ | ○ | Proceedings Paper | An Output-Capacitor-Free Adaptive-Frequency Digital LDO with a 420-mA Load Current and a Fast Settling Time | A typical digital low-dropout regulator (DLDO) that is clocked with a fixed frequency suffers from the trade-offs between power, speed, and stability. This paper proposes a DLDO that achieves both fast settling and low power consumption without limit-cycle oscillation (LCO) by adaptively changing the clock frequency and eventually turning off the clock in the steady state. The proposed LDO also features an inverter-based droopcompensation circuit for output-capacitor-free operation. The proposed LDO designed in a 28-nm CMOS process achieves a 70-ns settling time and a 137-mV droop voltage for the 400-mA load current transition with a 2.6-ns edge time, which translates to the figure of merit of 4.8 fs. | Min, Dong-Jick; Lee, Jun-Gi; Cho, Kunhee; Shim, Jae Hoon | Kyungpook Natl Univ, Sch Elect & Elect Engn, Daegu, South Korea | 57207202364; 58532344900; 57069360800; 56245719400 | alsehdwlr@knu.ac.kr;wnsakdhkd@knu.ac.kr;kunhee@knu.ac.kr;jhshim@knu.ac.kr; | 2023 IEEE INTERNATIONAL SYMPOSIUM ON CIRCUITS AND SYSTEMS, ISCAS | 0271-4302 | 0.61 | 2025-06-25 | 0 | 1 | Digital low dropout (DLDO) regulator; adaptive frequency; inverter-based droop-compensation; voltage-controlled oscillator (VCO); power management; linear regulator | LOW-DROPOUT REGULATOR; VOLTAGE DOWN-CONVERTER; LOW-QUIESCENT CURRENT | adaptive frequency; Digital low dropout (DLDO) regulator; inverter-based droop-compensation; linear regulator; power management; voltage-controlled oscillator (VCO) | Circuit oscillations; Economic and social effects; Electric current regulators; Electric inverters; Oscillistors; Power management; Variable frequency oscillators; Voltage regulators; Adaptive frequency; Digital low dropout regulator; Droop compensation; Inverte-based droop-compensation; Inverter-based; Linear regulators; Low dropout regulator; Output capacitor; Voltage-controled oscillator; Voltage-controlled-oscillator; Clocks | English | 2023 | 2023 | 10.1109/iscas46773.2023.10182029 | 바로가기 | 바로가기 | 바로가기 | ||||||||||||
| ○ | ○ | Proceedings Paper | An Overview of 3GPP Release 17 & 18 Advancements in the Context of V2X Technology | The 3rd Generation Partnership Project (3GPP) for the first time introduced the NR-V2X standard in its Rel.16, which is progressively improved by the researchers and industrial verticals to provide ubiquitous connectivity to the Internet-of-Vehicles (IoV). The 3GPP recently finalized NR-V2X standardization in its Rel.17. This paper overviews the recent technological advancements made in Rel. 17. Moreover, this paper summarizes the work items proposed for the 3GPP Rel. 18 and identified the potential research opportunities. The work items included in 3GPP Rel. 18 are primarily focused on integrating artificial intelligence/machine learning (AI/ML) into the core radio access network (C-RAN) to boost network performance. The 3GPP plenary consensus on Rel. 18 as an evolution to 5G-Advanced. This paper presents in-depth insights related to NR-V2X and its evolution to 5G-Advanced V2X. Energy efficiency, beamforming management, relaying User Equipment (UE), sidelink innovations, and Reduced device Capabilities (RedCAP) are the key aspects of the 5G-Advanced V2X. | Saad, Malik Muhammad; Tariq, Muhammad Ashar; Seo, Junho; Kim, Dongkyun | Kyungpook Natl Univ, Sch Comp Sci & Engn, Sangju Si, South Korea | Saad, Malik/ABF-9433-2021 | 57220715290; 57219865336; 57208740581; 35753648800 | maliksaad@knu.ac.kr;tariqashar@knu.ac.kr;jhseo@knu.ac.kr;dongkyun@knu.ac.kr; | 2023 INTERNATIONAL CONFERENCE ON ARTIFICIAL INTELLIGENCE IN INFORMATION AND COMMUNICATION, ICAIIC | 2831-6991 | 6.78 | 2025-06-25 | 13 | 15 | NR-V2X; 5G-Advanced; Re-evaluation mechanism; NR-V2X Mode 2; Resource Allocation | 5G-Advanced; NR-V2X; NR-V2X Mode 2; Re-evaluation mechanism; Resource Allocation | 5G mobile communication systems; Energy efficiency; Vehicle to Everything; 3rd generation; 5g-advanced; NR-V2X; NR-V2X mode 2; Potential researches; Re-evaluation; Re-evaluation mechanism; Resources allocation; Technological advancement; Work items; Radio access networks | English | 2023 | 2023 | 10.1109/icaiic57133.2023.10067121 | 바로가기 | 바로가기 | 바로가기 | ||||||||||||
| ○ | Article | Analysis and Stability Test of the Water Extracts and Powders from Piperis Longi Fructus for Toxicity Study | For toxicity assessment, powders and water extracts from the Piperis Longi Fructus were manufactured, and the component analysis were analyzed according to the Korean Herbal Pharmacopoeia method. Stability tests of powders and water extracts from the Piperis Longi Fructus were investigated at refrigerated and room temperature for one year. Powder and water extract were found to have been stable for a year after analyzing them with HPLC. Therefore, the use of the powders and water extracts of Piperis Longi Fructus after preparation during the animal test turned out to be stable. © 2023, Korean Society of Pharmacognosy. All rights reserved. | Hoang, Le Minh; An, Si Hyoung; Seok, Eun Yeong; Han, Kang-Hyun; Kim, Yong-Bum; Kim, Jeong Ah; Min, Byung Sun | College of Pharmacy, Drug Research and Development Center, Catholic University of Daegu, Gyeongbuk, Gyeongsan, 38430, South Korea; College of Pharmacy, Drug Research and Development Center, Catholic University of Daegu, Gyeongbuk, Gyeongsan, 38430, South Korea; College of Pharmacy, Drug Research and Development Center, Catholic University of Daegu, Gyeongbuk, Gyeongsan, 38430, South Korea; Korea Institute of Toxicology, Daejeon, 34114, South Korea; Korea Institute of Toxicology, Daejeon, 34114, South Korea; College of Pharmacy, Research Institute of Pharmaceutical Sciences, Kyungpook National University, Daegu, 41566, South Korea; College of Pharmacy, Drug Research and Development Center, Catholic University of Daegu, Gyeongbuk, Gyeongsan, 38430, South Korea | 59888078300; 58670079200; 58669942800; 57197058894; 55801563800; 24722570900; 55663638200 | jkim6923@knu.ac.kr;bsmin@cu.ac.kr; | Korean Journal of Pharmacognosy | 0253-3073 | 54 | 3 | 0 | 2025-06-25 | 0 | HPLC; Piperis Longi Fructus; Powder; Stability test; Water-extract | aqueous solution; article; fruit; human; medical literature; powder; room temperature | Korean | Final | 2023 | 10.22889/kjp.2023.54.3.137 | 바로가기 | 바로가기 |
페이지 이동: