연구성과로 돌아가기
2023 연구성과 (132 / 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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| ○ | ○ | Article | Numerical Analysis via Mixed Inverse Hydrodynamic Lubrication Theory of Reciprocating Rubber Seal Considering the Friction Thermal Effect | This study investigates how operating conditions such as ambient temperature and sealing pressure affect sealing performance for a typical U-cup seal. The developed analysis method combines inverse fluid lubrication (IHL) theory and the Greenwood-Williamson contact model (G-W model), and the effect of increasing surface temperature due to frictional heat generated between two surfaces is considered. Commercial FE software (ABAQUS) was used to simulate the interference fit analysis of rubber seals and the pressurized process. Through this model, the film distribution, working fluid leakage, and friction force in the sealing area were discussed according to the operating parameters, such as sealed pressure, rod velocity, and ambient temperature. The simulation results demonstrate the effect of fluid viscosity on oil film formation (which varies with ambient temperature), the effect of increasing the surface temperature, and the effect of surface roughness at a very small film thickness. | Kim, Bongjun; Suh, Junho; Lee, Bora; Chun, Yondo; Hong, Geuntae; Park, Jungjoon; Yu, Yonghun | Hyundai Heavy Ind, Ulsan 44032, South Korea; Pusan Natl Univ, Sch Mech Engn, Busan 46241, South Korea; Korea Electrotechnol Res Inst, Elect Machines & Drives Syst Res Ctr, Chang Won 51543, South Korea; Kyungpook Natl Univ, Dept Civil Engn, Daegu 41566, South Korea; Korea Railroad Res Inst, Innovat Transportat & Logist Res Ctr, Uiwang 16105, South Korea | ; Suh, Junho/LZF-8236-2025; Hong, Geuntae/AAM-8586-2020; YU, YONGHUN/AHB-1737-2022 | 58045819300; 56134393000; 56454553300; 7202638287; 57193714827; 58045524800; 57209560399 | yonghunyu87@krri.re.kr; | APPLIED SCIENCES-BASEL | APPL SCI-BASEL | 2076-3417 | 13 | 1 | SCIE | CHEMISTRY, MULTIDISCIPLINARY;ENGINEERING, MULTIDISCIPLINARY;MATERIALS SCIENCE, MULTIDISCIPLINARY;PHYSICS, APPLIED | 2023 | 2.5 | 24.0 | 0.26 | 2025-06-25 | 2 | 3 | friction; inverse hydrodynamic lubrication; leakage; mixed lubrication; reciprocating seal | RECTANGULAR ELASTOMERIC SEALS; ELASTOHYDRODYNAMIC LUBRICATION; O-RING; MODEL; LEAKAGE | friction; inverse hydrodynamic lubrication; leakage; mixed lubrication; reciprocating seal | English | 2023 | 2023-01 | 10.3390/app13010153 | 바로가기 | 바로가기 | 바로가기 | 바로가기 | |||
| ○ | ○ | Article | One-Step Gait Pattern Analysis of Hip Osteoarthritis Patients Based on Dynamic Time Warping through Ground Reaction Force | Osteoarthritis (OA) of the hip is a degenerative joint disease, which means it causes gradual damage to the joint, and its incidence rate continues to increase worldwide. Degenerative osteoarthritis can cause significant pain and gait disturbance in walking, affecting daily life. A diagnosis method for hip OA includes questioning and various walking movements to find abnormalities of gait patterns based on human observation. However, when multiple gait tests are performed to notice the gait, it can cause pain continuously, even during the examination. Suppose hip OA could be diagnosed with only a one-step gait; both patients and medical doctors would be benefited because the diagnosis time can be reduced and the burden on the patient is decreased dramatically. Therefore, in this paper, we aimed to propose a method to recognize the abnormality of the hip OA patient with a one-step gait pattern based on a dynamic time warping (DTW) algorithm through three directional ground reaction forces (GRFs). After a force plate measured three directional GRFs, the data of twenty-three hip OA patients and eighteen healthy people were classified using supervised machine learning algorithms. The results of the classification showed high accuracy and reliability. Then, the DTW algorithm was applied to compare the data of patients and healthy people to find out when patients may feel pain during the gait. By applying the DTW algorithm, it was possible to find out in which gait phase the patient's gait showed the difference, such as when the heel first contacted the ground, in the middle of walking, or when the toe came off the ground. Through the results, the data of the one-step gait on the force plate enabled us to classify patients and healthy people with a high accuracy of over 70%, recognize the abnormal gait pattern, and determine how to relieve the pain during the gait. | Ahn, Sohyun; Choi, Wiha; Jeong, Hieyong; Oh, Sehoon; Jung, Tae-Du | Chonnam Natl Univ, Dept Artificial Intelligence Convergence, 77 Yongbongro, Gwangju 61186, South Korea; Daegu Gyeongbuk Inst Sci & Technol DGIST, Dept Robot & Mechatron Engn, 333 Techno Jungang-Daero, Daegu 42988, South Korea; Kyungpook Natl Univ, Sch Med, 680 gukchaebosang-ro, Daegu 41404, South Korea | Oh, Sehoon/G-9312-2018; Oh, Sehoon/AAC-8132-2020 | 57278757500; 57190737827; 55461537500; 8709753900; 36622364500 | 217558@jnu.ac.kr;choiwiha@dgist.ac.kr;h.jeong@jnu.ac.kr;sehoon@dgist.ac.kr;teeed0522@knu.ac.kr; | APPLIED SCIENCES-BASEL | APPL SCI-BASEL | 2076-3417 | 13 | 8 | SCIE | CHEMISTRY, MULTIDISCIPLINARY;ENGINEERING, MULTIDISCIPLINARY;MATERIALS SCIENCE, MULTIDISCIPLINARY;PHYSICS, APPLIED | 2023 | 2.5 | 24.0 | 0.13 | 2025-06-25 | 1 | 1 | abnormal detection; dynamic time warping; hip osteoarthritis; one-step gait pattern | DISCRIMINANT-ANALYSIS; RECOMMENDATIONS | abnormal detection; dynamic time warping; hip osteoarthritis; one-step gait pattern | English | 2023 | 2023-04 | 10.3390/app13084665 | 바로가기 | 바로가기 | 바로가기 | 바로가기 | |||
| ○ | ○ | Article | Optimization of Design Parameters Using SQP for an Agricultural Pipe Extraction Device | Removal of agricultural pipes used in crop support and greenhouse agriculture is manpower-intensive. However, most agricultural workers are elderly. Therefore, auxiliary devices should be used to allow pipe removal with as little force as possible. In this study, the design parameters of an extraction device were optimized within constraints to minimize the force required to remove agricultural pipes. The optimization parameters are the length of each link and the initial link angle of the device. The driving force, according to the design parameters, was calculated by applying the theory of kinematics. An optimal design plan was derived using an optimization algorithm to minimize the force driving the device within the desired constraint. As a result of the optimization, it was confirmed that the force required to remove the pipe was reduced by 87.1% compared with the initial design value and was designed to suit the user's convenience. | Lee, Su-Min; Lee, Sang-Hong; Han, Hyun-Woo; Oh, Jooseon; Shim, Sung-Bo | Kyungpook Natl Univ, Dept Bioind Engn, Daegu 37224, South Korea; Seoul Natl Univ, Dept Biosyst Engn, Seoul 08826, South Korea; Chonnam Natl Univ, Dept Convergence Biosyst Engn, Gwangju 61186, South Korea; Kyungpook Natl Univ, Smart Agr Innovat Ctr, Daegu 37224, South Korea; Kyungpook Natl Univ, Upland Field Machinery Res Ctr, Daegu 37224, South Korea | Oh, Jooseon/HJH-9372-2023 | 58137353700; 58137800400; 57211044219; 57208126881; 56482771400 | jooseon.oh@jnu.ac.kr;sbs80@knu.ac.kr; | APPLIED SCIENCES-BASEL | APPL SCI-BASEL | 2076-3417 | 13 | 5 | SCIE | CHEMISTRY, MULTIDISCIPLINARY;ENGINEERING, MULTIDISCIPLINARY;MATERIALS SCIENCE, MULTIDISCIPLINARY;PHYSICS, APPLIED | 2023 | 2.5 | 24.0 | 0 | 2025-06-25 | 0 | 0 | agricultural pipe; optimum design; sequential quadratic programming; EMG value | agricultural pipe; EMG value; optimum design; sequential quadratic programming | English | 2023 | 2023-03 | 10.3390/app13053167 | 바로가기 | 바로가기 | 바로가기 | 바로가기 | ||||
| ○ | ○ | Article | Relative Pose Estimation between Image Object and ShapeNet CAD Model for Automatic 4-DoF Annotation | Estimating the three-dimensional (3D) pose of real objects using only a single RGB image is an interesting and difficult topic. This study proposes a new pipeline to estimate and represent the pose of an object in an RGB image only with the 4-DoF annotation to a matching CAD model. The proposed method retrieves CAD candidates from the ShapeNet dataset and utilizes the pose-constrained 2D renderings of the candidates to find the best matching CAD model. The pose estimation pipeline consists of several steps of learned networks followed by image similarity measurements. First, from a single RGB image, the category and the object region are determined and segmented. Second, the 3-DoF rotational pose of the object is estimated by a learned pose-contrast network only using the segmented object region. Thus, 2D rendering images of CAD candidates are generated based on the rotational pose result. Finally, an image similarity measurement is performed to find the best matching CAD model and to determine the 1-DoF focal length of the camera to align the model with the object. Conventional pose estimation methods employ the 9-DoF pose parameters due to the unknown scale of both image object and CAD model. However, this study shows that only 4-DoF annotation parameters between real object and CAD model is enough to facilitates the projection of the CAD model to the RGB space for image-graphic applications such as Extended Reality. In the experiments, performance of the proposed method is analyzed by using ground truth and comparing with a triplet-loss learning method. | Park, Soon-Yong; Son, Chang-Min; Jeong, Won-Jae; Park, Sieun | Kyungpook Natl Univ, Sch Elect Engn, Daegu 41566, South Korea; Kyungpook Natl Univ, Grad Sch Elect & Elect Engn, Daegu 41566, South Korea | ; Park, Soon-Yong/HGV-2374-2022 | 7501834063; 57826492700; 59270595000; 59685274700 | sypark@knu.ac.kr; | APPLIED SCIENCES-BASEL | APPL SCI-BASEL | 2076-3417 | 13 | 2 | SCIE | CHEMISTRY, MULTIDISCIPLINARY;ENGINEERING, MULTIDISCIPLINARY;MATERIALS SCIENCE, MULTIDISCIPLINARY;PHYSICS, APPLIED | 2023 | 2.5 | 24.0 | 0.39 | 2025-06-25 | 2 | 3 | pose estimation; CAD retrieval; ShapeNet; image similarity; 4-DoF annotation; extended reality | 4-DoF annotation; CAD retrieval; extended reality; image similarity; pose estimation; ShapeNet | English | 2023 | 2023-01 | 10.3390/app13020693 | 바로가기 | 바로가기 | 바로가기 | 바로가기 | ||||
| ○ | ○ | Article | Requirements for Dental CAD Software: A Survey of Korean Dental Personnel | This study aimed to evaluate the needs of dentists, dental technicians, dental hygienists, and dental-related workers in terms of dental computer-aided design (CAD) software and artificial intelligence (AI). Based on a factor analysis, 57 survey items were assigned to six categories: (a) considerations when purchasing dental CAD software; (b) prosthesis design process; (c) dental CAD function; (d) use of AI dental CAD crown and denture design; (e) impact of AI; and (f) improvements in AI features. Overall, 93 participants were included in the study, and the reliability of the resultant survey data was estimated using Cronbach's alpha coefficient. Statistically significant factors in each category were identified using one-way analysis of variance and Tukey's honestly significant difference test (alpha = 0.05). The results revealed that design quality, design convenience and reproducibility, margin line setting, and automatic margin setting were considered most important in their respective categories (p < 0.05). There were also significant differences in the importance of certain items, such as branding importance and functional diversity, among different dental personnel groups (p < 0.05). Design speed and convenience were also found to be more important to dentists and dental hygienists compared to other dental personnel (p < 0.05). The importance of specific survey items varied significantly based on age, dental personnel, and work experience groups. Dental personnel, such as dentists and dental technicians, assigned greater importance to certain factors, such as branding, functional diversity, design speed, and compatibility with CAM equipment, compared to other occupational groups. | Son, KeunBaDa; Kim, Gyu Ri; Kim, Won-Gi; Kang, Wol; Lee, Du-Hyeong; Kim, So-Yeun; Lee, Jae-Mok; Kim, Yong-Gun; Kim, Jin-Wook; Lee, Sung-Tak; Jin, Myoung-Uk; Kim, Ho-Jin; Lee, Jaesik; Kim, Ji-Rak; Lee, Kyu-Bok | Kyungpook Natl Univ, Adv Dent Device Dev Inst A3DI, Daegu 41940, South Korea; Kyungpook Natl Univ, Grad Sch, Dept Dent Sci, Daegu 41940, South Korea; Daegu Hlth Coll, Dept Dent Technol, Daegu 41453, South Korea; Kyungpook Natl Univ, Sch Dent, Dept Prosthodont, Daegu 41940, South Korea; Kyungpook Natl Univ, Sch Dent, Dept Periodontol, Daegu 41940, South Korea; Kyungpook Natl Univ, Sch Dent, Dept Oral & Maxillofacial Surg, Daegu 41940, South Korea; Kyungpook Natl Univ, Sch Dent, Dept Conservat Dent, Daegu 41940, South Korea; Kyungpook Natl Univ, Sch Dent, Dept Orthodont, Daegu 41940, South Korea; Kyungpook Natl Univ, Sch Dent, Dept Pediat Dent, Daegu 41940, South Korea; Kyungpook Natl Univ, Sch Dent, Dept Oral Med, Daegu 41940, South Korea | ; SON, Keunbada/AAG-8089-2019; Son, Keunbada/AAG-8089-2019 | 57202916520; 57222585103; 34770168500; 59586027900; 35770948000; 57190972249; 17346330000; 55622694400; 55862646000; 55931708300; 56492091300; 57200084686; 57193887436; 57207438323; 15925571200 | kblee@knu.ac.kr; | APPLIED SCIENCES-BASEL | APPL SCI-BASEL | 2076-3417 | 13 | 5 | SCIE | CHEMISTRY, MULTIDISCIPLINARY;ENGINEERING, MULTIDISCIPLINARY;MATERIALS SCIENCE, MULTIDISCIPLINARY;PHYSICS, APPLIED | 2023 | 2.5 | 24.0 | 0.52 | 2025-06-25 | 3 | 4 | dentistry; dental CAD software; survey; artificial intelligence | CROWNS; TRUENESS; DESIGN; FIT | artificial intelligence; dental CAD software; dentistry; survey | English | 2023 | 2023-03 | 10.3390/app13052803 | 바로가기 | 바로가기 | 바로가기 | 바로가기 | |||
| ○ | Article | Synthesis of Some Fluorescent Dyes Based on Stilbene Derivatives with Various Substituents and Their Effects on the Absorption Maxima | The six stilbene-based dyes containing benzoxazole substituents to improve solubility of dyes as well as the efficiency of fluorescence at blue emission were synthesized. In this work, absorption and fluorescent properties of the synthesized dyes were investigated. For the derivatization of benzoxazolyl stilbene dye, -NO2 and -NH2 groups were introduced in sequence onto benzoxazolyl rings. The emission maxima of the six dyes prepared were observed in the range of 435 nm similar to 471 nm. In addition, the solubility of the dyes in dichloromethane was examined for application to the nonpolar polymer films such as PE, PP, PVC and so on. N-alkyl groups were determined to have a greater solubility of alkylated stilbene-based dyes than analogue containing and unsubstituted group. Furthermore, investigation of the optical effects of tortional strain according to conformation of side group was also performed. For identifying these properties, the geometry, dihedral angle, and other parameters of synthesized dyes were calculated by the density functional theory and time-dependent function using a gaussian 09 program. | Lee, Yoon-Gu; Choi, Jae-Hong | Kyungpook Natl Univ, Dept Text Syst Engn, 80 DaeHak Ro, Daegu 41566, South Korea | jaehong@knu.ac.kr; | APPLIED SCIENCES-BASEL | APPL SCI-BASEL | 2076-3417 | 13 | 9 | SCIE | CHEMISTRY, MULTIDISCIPLINARY;ENGINEERING, MULTIDISCIPLINARY;MATERIALS SCIENCE, MULTIDISCIPLINARY;PHYSICS, APPLIED | 2023 | 2.5 | 24.0 | 6 | blue fluorescence; stilbene; benzoxazole; emission maxima; absorption maxima; dihedral angle | ALKYL CHAIN-LENGTH; QUANTUM YIELD; BIS(BENZOXAZOLYL)STILBENE; AGGREGATION; TEMPERATURE; PERFORMANCE; TRIAZINE | English | 2023 | 2023-04-29 | 10.3390/app13095543 | 바로가기 | 바로가기 | 바로가기 | |||||||||||
| ○ | ○ | Article | Target-Aware Feature Bottleneck for Real-Time Visual Tracking | Recent Siamese network-based visual tracking approaches have achieved high performance metrics on numerous recent visual tracking benchmarks, where most of these trackers employ a backbone feature extractor network with a prediction head network for classification and regression tasks. However, there has been a constant trend of employing a larger and complex backbone network and prediction head networks for improved performance, where increased computational load can slow down the overall speed of the tracking algorithm. To address the aforementioned issues, we propose a novel target-aware feature bottleneck module for trackers, where the proposed bottleneck can elicit a target-aware feature in order to obtain a compact feature representation from the backbone network for improved speed and robustness. Our lightweight target-aware bottleneck module attends to the feature representation of the target region to elicit scene-specific information and generate feature-wise modulation weights that can adaptively change the importance of each feature. The proposed tracker is evaluated on large-scale visual tracking datasets, GOT-10k and LaSOT, and we achieve real-time speed in terms of computation and obtain improved accuracy over the baseline tracker algorithm with high performance metrics. | Choi, Janghoon | Kyungpook Natl Univ, Grad Sch Data Sci, Daegu 41566, South Korea | 57202773325 | jhchoi09@knu.ac.kr; | APPLIED SCIENCES-BASEL | APPL SCI-BASEL | 2076-3417 | 13 | 18 | SCIE | CHEMISTRY, MULTIDISCIPLINARY;ENGINEERING, MULTIDISCIPLINARY;MATERIALS SCIENCE, MULTIDISCIPLINARY;PHYSICS, APPLIED | 2023 | 2.5 | 24.0 | 0.13 | 2025-06-25 | 1 | 1 | visual tracking; model-free tracking; object tracking; bottleneck module; real-time tracking | bottleneck module; model-free tracking; object tracking; real-time tracking; visual tracking | English | 2023 | 2023-09 | 10.3390/app131810198 | 바로가기 | 바로가기 | 바로가기 | 바로가기 | |||||
| ○ | ○ | Article | Time Efficiency Improvement in Quadruped Walking with Supervised Training Joint Model | To generate stable walking of a quadruped, the complexity of the configuration of the robot involves a significant amount of optimization that decreases to its time efficiency. To address this issue, a machine learning method was used to build a simplified control policy using joint models for the supervised training of quadruped robots. This study considered 12 joints for a four-legged robot, and each joint value was determined based on the conventional method of walking simulation and prepossessed, equaling 2508 sets of data. For data training, the multilayer perceptron model was used, and the optimized number of epochs used to train the model was 5000. The trained models were implemented in robot walking simulations, and they improved performance with an average distance error of 0.0719 m and a computational time as low as 91.98 s. | Yeoh, Chin Ean; Ahn, Min Sung; Choi, Soomin; Yi, Hak | Kyungpook Natl Univ, Grad Sch, Dept Mech Engn, Daegu 41566, South Korea; Univ Calif Los Angeles, Dept Mech & Aerosp Engn, 420 Westwood Plaza,Rm 32-117E, Los Angeles, CA 90095 USA | Yeoh, Chin Ean/JRY-5058-2023 | 57221051109; 57061824700; 56124305600; 56567311000 | schoi@knu.ac.kr;yihak@knu.ac.kr; | APPLIED SCIENCES-BASEL | APPL SCI-BASEL | 2076-3417 | 13 | 4 | SCIE | CHEMISTRY, MULTIDISCIPLINARY;ENGINEERING, MULTIDISCIPLINARY;MATERIALS SCIENCE, MULTIDISCIPLINARY;PHYSICS, APPLIED | 2023 | 2.5 | 24.0 | 0 | 2025-06-25 | 0 | 0 | supervised learning; quadruped robot; walking locomotion; multilayer perceptron | multilayer perceptron; quadruped robot; supervised learning; walking locomotion | English | 2023 | 2023-02 | 10.3390/app13042658 | 바로가기 | 바로가기 | 바로가기 | 바로가기 | ||||
| ○ | ○ | Article | Uncertainty Quantification of Ride Comfort Based on gPC Framework for a Fully Coupled Human-Vehicle Model | We investigated the stochastic response of a person sitting in a driving vehicle to quantify the impact of an uncertain parameter important in controlling defect reduction in terms of ride comfort. Using CarSim software and MATLAB/Simulink, we developed a fully coupled model that simulates a driving vehicle combined with an analytical nonlinear human model. Ride comfort was evaluated as a ride index considering the frequency weights defined in BS 6841. Additionally, to investigate the uncertainty of the ride index, a framework for calculating the ride index was proposed using the generalized polynomial (gPC) method. Further, sensitivity analysis of the ride index was performed for each uncertainty parameter, such as stiffness and damping. The results obtained through the gPC method were in good agreement with those obtained via Monte Carlo simulation (MCS) and were excellent in terms of computation time without a loss of numerical accuracy. Through in-depth investigation, we found that the stochastic distribution of the ride index varies differently for each uncertain parameter in the human model. By comparing linear and nonlinear human models, we also found that the nonlinearity of the human model is an important concern in the stochastic estimation of ride comfort. | Song, Byoung-Gyu; Bae, Jong-Jin; Kang, Namcheol | Kyungpook Natl Univ, Sch Mech Engn, Daegu 41566, South Korea; Korea Aerosp Res Inst, KSLV II R&D Directorate, Daejeon 34133, South Korea | 57211027939; 55631622800; 24830970900 | sbg045@knu.ac.kr;jjbae@kari.re.kr;nckang@knu.ac.kr; | APPLIED SCIENCES-BASEL | APPL SCI-BASEL | 2076-3417 | 13 | 11 | SCIE | CHEMISTRY, MULTIDISCIPLINARY;ENGINEERING, MULTIDISCIPLINARY;MATERIALS SCIENCE, MULTIDISCIPLINARY;PHYSICS, APPLIED | 2023 | 2.5 | 24.0 | 0.26 | 2025-06-25 | 2 | 2 | human-vehicle model; uncertainty; ride index; generalized polynomial chaos; stochastic analysis | POLYNOMIAL CHAOS; MULTIBODY SYSTEMS; VIBRATION; DYNAMICS | generalized polynomial chaos; human–vehicle model; ride index; stochastic analysis; uncertainty | English | 2023 | 2023-06-02 | 10.3390/app13116785 | 바로가기 | 바로가기 | 바로가기 | 바로가기 | ||||
| ○ | ○ | Article | Vision-Based Activity Classification of Excavators by Bidirectional LSTM | Advancements in deep learning and vision-based activity recognition development have significantly improved the safety, continuous monitoring, productivity, and cost of the earthwork site. The construction industry has adopted the CNN and RNN models to classify the different activities of construction equipment and automate the construction operations. However, the currently available methods in the industry classify the activities based on the visual information of current frames. To date, the adjacent visual information of current frames has not been simultaneously examined to recognize the activity in the construction industry. This paper proposes a novel methodology to classify the activities of the excavator by processing the visual information of video frames adjacent to the current frame. This paper follows the CNN-BiLSTM standard deep learning pipeline for excavator activity recognition. First, the pre-trained CNN model extracted the sequential pattern of visual features from the video frames. Then BiLSTM classified the different activities of the excavator by analyzing the output of the pre-trained convolutional neural network. The forward and backward LSTM layers stacked on help the algorithm compute the output by considering previous and upcoming frames' visual information. Experimental results have shown the average precision and recall to be 87.5% and 88.52%, respectively. | Kim, In-Sup; Latif, Kamran; Kim, Jeonghwan; Sharafat, Abubakar; Lee, Dong-Eun; Seo, Jongwon | Hanyang Univ, Dept Civil & Environm Engn, Seoul 04763, South Korea; Korea Natl Univ Transportat, Dept Civil Engn, Chungbuk 27469, South Korea; Kyungpook Natl Univ, Sch Architectural Civil Environm & Energy Engn, Daegu 41566, South Korea | Sharafat, Abubakar/ITW-2048-2023 | 57798279200; 57221094832; 55720258400; 57204630290; 56605563300; 7401783784 | dolee@knu.ac.kr;jseo@hanyang.ac.kr; | APPLIED SCIENCES-BASEL | APPL SCI-BASEL | 2076-3417 | 13 | 1 | SCIE | CHEMISTRY, MULTIDISCIPLINARY;ENGINEERING, MULTIDISCIPLINARY;MATERIALS SCIENCE, MULTIDISCIPLINARY;PHYSICS, APPLIED | 2023 | 2.5 | 24.0 | 1.95 | 2025-06-25 | 15 | 15 | computer vision; activity recognition; convolution neural network (CNN); long short-term memory (LSTM); Googlenet; visual features | CONVOLUTIONAL NEURAL-NETWORK; ACTION RECOGNITION; EARTHMOVING EXCAVATORS; CONSTRUCTION WORKERS; EQUIPMENT; MODEL; WORKFORCE; FEATURES; TIME | activity recognition; computer vision; convolution neural network (CNN); Googlenet; long short-term memory (LSTM); visual features | English | 2023 | 2023-01 | 10.3390/app13010272 | 바로가기 | 바로가기 | 바로가기 | 바로가기 | |||
| ○ | ○ | Article | Visual Odometry of a Low-Profile Pallet Robot Based on Ortho-Rectified Ground Plane Image from Fisheye Camera | This study presents a visual-only odometry technique of a low-profile pallet robot using image feature tracking in ground plane images generated from a fisheye camera. The fisheye camera is commonly used in many robot vision applications because it provides a larger field of view (FoV) around a robot. However, because of the large radial distortion, the fisheye image is generally converted to a pinhole image for visual feature tracking or matching. Although the radial distortion can be eliminated via image undistortion with the lens calibration parameters, it causes several side effects, such as degraded image resolution and a significant reduction in the FoV. In this paper, instead of using the pinhole model, we propose to generate a ground plane image (GPI) from the fisheye image. GPI is a virtual top-view image that only contains the ground plane at the front of the robot. First, the original fisheye image is projected to several virtual pinhole images to generate a cubemap. Second, the front and bottom faces of the cubemap are projected to a GPI. Third, the GPI is homographically transformed again to further reduce image distortion. As a result, an accurate ortho-rectified ground plane image is obtained from the virtual top-view camera. For visual odometry using the ortho-rectified GPI, a number of 2D motion vectors are obtained using feature extraction and tracking between the previous and current frames in the GPI. By calculating a scaled motion vector, which is the measurement of the virtual wheel encoder of the mobile robot, we estimate the velocity and steering angle of the virtual wheel using the motion vector. Finally, we estimate the pose of the mobile robot by applying a kinematic model to the mobile robot. | Park, Soon-Yong; Lee, Ung-Gyo; Baek, Seung-Hae | Kyungpook Natl Univ, Sch Elect & Elect Engn, 80 Daehak Ro, Daegu 41566, South Korea; HL Klemove, 224 Harmony Ro, Incheon 22011, South Korea; KLA Corp, 830 Dongtansunwhan Daero, Hwaseong 18468, South Korea | Park, Soon-Yong/HGV-2374-2022 | 7501834063; 57270887000; 48861235500 | sypark@knu.ac.kr;dndry123@naver.com;eardrop77@gmail.com; | APPLIED SCIENCES-BASEL | APPL SCI-BASEL | 2076-3417 | 13 | 16 | SCIE | CHEMISTRY, MULTIDISCIPLINARY;ENGINEERING, MULTIDISCIPLINARY;MATERIALS SCIENCE, MULTIDISCIPLINARY;PHYSICS, APPLIED | 2023 | 2.5 | 24.0 | 0.13 | 2025-06-25 | 1 | 1 | fisheye lens; cubemap; ground plane image; pallet mobile robot | cubemap; fisheye lens; ground plane image; pallet mobile robot | English | 2023 | 2023-08 | 10.3390/app13169095 | 바로가기 | 바로가기 | 바로가기 | 바로가기 | ||||
| ○ | ○ | Article | Weakly Supervised U-Net with Limited Upsampling for Sound Event Detection | Sound event detection (SED) is the task of finding the identities of sound events, as well as their onset and offset timings from audio recordings. When complete timing information is not available in the training data, but only the event identities are known, SED should be solved by weakly supervised learning. The conventional U-Net with global weighted rank pooling (GWRP) has shown a decent performance, but extensive computation is demanded. We propose a novel U-Net with limited upsampling (LUU-Net) and global threshold average pooling (GTAP) to reduce the model size, as well as the computational overhead. The expansion along the frequency axis in the U-Net decoder was minimized, so that the output map sizes were reduced by 40% at the convolutional layers and 12.5% at the fully connected layers without SED performance degradation. The experimental results on a mixed dataset of DCASE 2018 Tasks 1 and 2 showed that our limited upsampling U-Net (LUU-Net) with GTAP was about 23% faster in training and achieved 0.644 in audio tagging and 0.531 in weakly supervised SED tasks in terms of F1 scores, while U-Net with GWRP showed 0.629 and 0.492, respectively. The major contribution of the proposed LUU-Net is the reduction in the computation time with the SED performance being maintained or improved. The other proposed method, GTAP, further improved the training time reduction and provides versatility for various audio mixing conditions by adjusting a single hyperparameter. | Lee, Sangwon; Kim, Hyemi; Jang, Gil-Jin | Kyungpook Natl Univ, Sch Elect & Elect Engn, Daegu 41566, South Korea; Elect & Telecommun Res Inst, Daejeon 34129, South Korea | 59286017000; 56981395100; 7102646102 | gjang@knu.ac.kr; | APPLIED SCIENCES-BASEL | APPL SCI-BASEL | 2076-3417 | 13 | 11 | SCIE | CHEMISTRY, MULTIDISCIPLINARY;ENGINEERING, MULTIDISCIPLINARY;MATERIALS SCIENCE, MULTIDISCIPLINARY;PHYSICS, APPLIED | 2023 | 2.5 | 24.0 | 0.39 | 2025-06-25 | 1 | 3 | sound event detection; U-Net; weakly supervised learning; pooling | pooling; sound event detection; U-Net; weakly supervised learning | English | 2023 | 2023-06-04 | 10.3390/app13116822 | 바로가기 | 바로가기 | 바로가기 | 바로가기 | |||||
| ○ | ○ | Article | Zero-Shot Image Classification with Rectified Embedding Vectors Using a Caption Generator | Although image recognition technologies are developing rapidly with deep learning, conventional recognition models trained by supervised learning with class labels do not work well when test inputs from untrained classes are given. For example, a recognizer trained to classify Asian bird species cannot recognize the species of kiwi, because the class label "kiwi" and its image samples have not been seen during training. To overcome this limitation, zero-shot classification has been studied recently, and the joint-embedding-based approach has been suggested as one of the promised solutions. In this approach, image features and text descriptions belonging to the same class are trained to be closely located in a common joint-embedding space. Once we obtain the embedding function that can represent the semantic relationship of image-text pairs in training data, test images and text descriptions (prototypes) of unseen classes can also be mapped to the joint-embedding space for classification. The main challenge with this approach is mapping inputs of two different modalities into a common space, and previous works suffer from the inconsistency between the distribution of two feature sets on joint-embedding space extracted from the heterogeneous inputs. To treat this problem, we propose a novel method of employing additional textual information to rectify the visual representation of input images. Since the conceptual information of test classes is generally given as texts, we expect that the additional descriptions from a caption generator can adjust the visual feature for better matching with the representation of the test classes. We also propose to use the generated textual descriptions to augment training samples for learning joint-embedding space. In the experiments on two benchmark datasets, the proposed method shows significant performance improvements of 1.4% on the CUB dataset and 5.5% on the flower dataset, in comparison to existing models. | Hur, Chan; Park, Hyeyoung | Kyungpook Natl Univ, Sch Comp Sci & Engn, Daegu 41566, South Korea | 57216416244; 55713613500 | chanhur@knu.ac.kr;hypark@knu.ac.kr; | APPLIED SCIENCES-BASEL | APPL SCI-BASEL | 2076-3417 | 13 | 12 | SCIE | CHEMISTRY, MULTIDISCIPLINARY;ENGINEERING, MULTIDISCIPLINARY;MATERIALS SCIENCE, MULTIDISCIPLINARY;PHYSICS, APPLIED | 2023 | 2.5 | 24.0 | 0.13 | 2025-06-25 | 1 | 1 | zero-shot learning; image captioning; joint-embedding; visual feature enhancement; textural feature generation | image captioning; joint-embedding; textural feature generation; visual feature enhancement; zero-shot learning | English | 2023 | 2023-06 | 10.3390/app13127071 | 바로가기 | 바로가기 | 바로가기 | 바로가기 | |||||
| ○ | ○ | Article | Annotation-Efficient Deep Learning Model for Pancreatic Cancer Diagnosis and Classification Using CT Images: A Retrospective Diagnostic Study | Simple Summary In computer-assisted diagnostics for pancreatic cancer, attributes featuring irregular contours and indistinct boundaries on CT images present challenges in acquiring high-quality annotations. In response to this issue, we have devised an innovative self-supervised learning algorithm, engineered to enhance the differentiation of malignant and benign lesions. This innovation obviates the necessity for radiologist intervention, thus facilitating the precise classification of pancreatic cancer. By employing a pseudo-lesion segmentation self-supervised learning model, which capitalizes on automatically generated high-quality training data, we have managed to significantly elevate the performance of both convolutional neural network-based and transformer-based deep learning models. The aim of this study was to develop a novel deep learning (DL) model without requiring large-annotated training datasets for detecting pancreatic cancer (PC) using computed tomography (CT) images. This retrospective diagnostic study was conducted using CT images collected from 2004 and 2019 from 4287 patients diagnosed with PC. We proposed a self-supervised learning algorithm (pseudo-lesion segmentation (PS)) for PC classification, which was trained with and without PS and validated on randomly divided training and validation sets. We further performed cross-racial external validation using open-access CT images from 361 patients. For internal validation, the accuracy and sensitivity for PC classification were 94.3% (92.8-95.4%) and 92.5% (90.0-94.4%), and 95.7% (94.5-96.7%) and 99.3 (98.4-99.7%) for the convolutional neural network (CNN) and transformer-based DL models (both with PS), respectively. Implementing PS on a small-sized training dataset (randomly sampled 10%) increased accuracy by 20.5% and sensitivity by 37.0%. For external validation, the accuracy and sensitivity were 82.5% (78.3-86.1%) and 81.7% (77.3-85.4%) and 87.8% (84.0-90.8%) and 86.5% (82.3-89.8%) for the CNN and transformer-based DL models (both with PS), respectively. PS self-supervised learning can increase DL-based PC classification performance, reliability, and robustness of the model for unseen, and even small, datasets. The proposed DL model is potentially useful for PC diagnosis. | Viriyasaranon, Thanaporn; Chun, Jung Won; Koh, Young Hwan; Cho, Jae Hee; Jung, Min Kyu; Kim, Seong-Hun; Kim, Hyo Jung; Lee, Woo Jin; Choi, Jang-Hwan; Woo, Sang Myung | Ewha Womans Univ, Grad Program Syst Hlth Sci & Engn, Div Mech & Biomed Engn, Seoul 03760, South Korea; Natl Canc Ctr, Ctr Liver & Pancreatobiliary Canc, Goyang 10408, South Korea; Yonsei Univ, Gangnam Severance Hosp, Dept Internal Med, Coll Med, Seoul 03722, South Korea; Kyungpook Natl Univ Hosp, Dept Internal Med, Daegu 41944, South Korea; Jeonbuk Natl Univ, Jeonbuk Natl Univ Hosp, Dept Internal Med, Res Inst Clin Med,Biomed Res Inst, Jeonju 54907, South Korea; Korea Univ, Dept Gastroenterol, Guro Hosp, Seoul 10408, South Korea | Kim, Young/J-5414-2012; Cho, Jae Hee/AAL-8192-2020 | 57193691180; 57211335949; 55539797900; 58838319500; 56783168100; 57206878084; 56995392900; 55619312784; 55850525400; 14038271300 | thanaporn.v@ewhain.net;deli4927@ncc.re.kr;mrikyh@ncc.re.kr;hcho9328@yuhs.ac;minky1973@knu.ac.kr;shkimgi@jbnu.ac.kr;hjkimmd@korea.ac.kr;lwj@ncc.re.kr;choij@ewha.ac.kr;wsm@ncc.re.kr; | CANCERS | CANCERS | 2072-6694 | 15 | 13 | SCIE | ONCOLOGY | 2023 | 4.5 | 24.1 | 1.51 | 2025-06-25 | 8 | 11 | classification; deep learning; diagnosis; medical imaging; pancreatic cancer | classification; deep learning; diagnosis; medical imaging; pancreatic cancer | adult; Article; cancer classification; cancer patient; computer assisted tomography; controlled study; convolutional neural network; deep learning; diagnostic accuracy; diagnostic value; human; image segmentation; instrument validation; intermethod comparison; major clinical study; measurement precision; middle aged; pancreas cancer; predictive value; pseudo lesion segmentation; reliability; residual neural network; retrospective study; sensitivity analysis; sensitivity and specificity; transformer based deep learning | English | 2023 | 2023-07 | 10.3390/cancers15133392 | 바로가기 | 바로가기 | 바로가기 | 바로가기 | |||
| ○ | ○ | Article | Examining Final-Administered Medication as a Measure of Data Quality: A Comparative Analysis of Death Data with the Central Cancer Registry in Republic of Korea | Simple Summary Death represents the definitive endpoint for a patient; therefore, it is crucial to determine an accurate date of death. This study aims to examine the final-administered medication in a gold standard cohort that assesses death data accuracy. By utilizing electronic health records from a single medical institution and the Korean Central Cancer Registry, we establish the gold standard as patients who died in the hospital after the implementation of electronic health records, with a difference of 0 or 1 day between the final hospital visit/discharge and death. We calculate the similarity of the terminal medication between the gold standard and cohorts using cosine similarity. The findings reveal a positive correlation between mortality rates and similarities of the final-administered medication. This study introduces the potential of the last administered medication as a novel data quality measure of death data when the date of death differs between datasets. Death is a crucial outcome in retrospective cohort studies, serving as a criterion for analyzing mortality in a database. This study aimed to assess the quality of extracted death data and investigate the potential of the final-administered medication as a variable to quantify accuracy for the validation dataset. Electronic health records from both an in-hospital and the Korean Central Cancer Registry were used for this study. The gold standard was established by examining the differences between the dates of in-hospital deaths and cancer-registered deaths. Cosine similarity was employed to quantify the final-administered medication similarities between the gold standard and other cohorts. The gold standard was determined as patients who died in the hospital after 2006 and whose final hospital visit/discharge date and death date differed by 0 or 1 day. For all three criteria-(a) cancer stage, (b) cancer type, and (c) type of final visit-there was a positive correlation between mortality rates and the similarities of the final-administered medication. This study introduces a measure that can provide additional accurate information regarding death and differentiates the reliability of the dataset. | Tak, Yae Won; Han, Jeong Hyun; Park, Yu Jin; Kim, Do-Hoon; Oh, Ji Seon; Lee, Yura | Univ Ulsan, Asan Med Ctr, Dept Informat Med, Coll Med, Seoul 05505, South Korea; Asan Med Ctr, Med Informat Management Team, Seoul 05505, South Korea; Kyungpook Natl Univ Hosp, Med Big Data Res Ctr, Daegu 41944, South Korea | 57776802600; 57777478700; 57944823500; 55624468392; 35590419200; 56609515600 | yaewon.c.tak@amc.seoul.kr;jhhan121@amc.seoul.kr;pyj81@amc.seoul.kr;k8016851@gmail.com;doogie55@naver.com;haepary@naver.com;haepary@amc.seoul.kr; | CANCERS | CANCERS | 2072-6694 | 15 | 13 | SCIE | ONCOLOGY | 2023 | 4.5 | 24.1 | 0.15 | 2025-06-25 | 1 | 1 | death; mortality; data quality; public data; data accuracy | RECORDS; STAGE | data accuracy; data quality; death; mortality; public data | adult; article; cancer patient; cancer registry; cancer staging; cohort analysis; controlled study; data accuracy; date of death; electronic health record; female; gold standard; hospital mortality; human; in-hospital mortality; male; mortality; mortality rate; quantitative analysis; reliability; retrospective study; South Korea | English | 2023 | 2023-07 | 10.3390/cancers15133371 | 바로가기 | 바로가기 | 바로가기 | 바로가기 |
페이지 이동: