• Publications . . . . 1 match
        1. Sanghyun Choo, '''Hoonseok Park''', Jae-Yoon Jung, Kevin Flores, and Chang S. Nam*, "Improving Classification Performance of Motor Imagery BCI through EEG Data Augmentation with Conditional Generative Adversarial Networks", ''Neural Network'', vol. 180, Dec 2024, 106665. (SCIE, IF=6.0, 10%, Neuro) [https://doi.org/10.1016/j.neunet.2024.106665 doi]
        1. '''Young Ha Joo''', '''Hoonseok Park''', Haejoong Kim, Ri Choe, Younkook Kang, and Jae-Yoon Jung*, "Traffic Flow Speed Prediction in Overhead Transport Systems for Semiconductor Fab Using Dense-UNet", ''Processes'', Vol. 10, No. 8, Aug. 2022, 1580. (SCIE, IF=3.352, 48.2%, Chem). [https://www.mdpi.com/2227-9717/10/8/1580 link] [https://www.mdpi.com/2227-9717/10/8/1580/pdf pdf]
• Seminars . . . . 16 matches
        * 주영하: UNet revision 논의, GCN 실험 논의
        * UNet : Processes (SCIE, mdpi)
        * 영하 : '''SS UNet의 미래는? ''' (국내 또는 mdpi?)
       UNet -> 방안 고민(??)
        * Papers : '''OHT-UNet 추가실험(by 2/20)->훈석''', GCN for OHT (학위논문)
        * 영하 - Logistics - UNet (CNN)
        * UNet - '''영하''', 훈석 (50분)
        * DenseUNet => Hyperparameter 목록 기재 => 하나씩 바꿔보면서 최적 세트 찾기 (w/ Jung)
        * Joo : 성능높이기(UNet Tuning, 설계 input 추가) : 4가지 조합 실험 (DenseUNet 유무 * 설계 유무) -> IEEE TSM
        * 주영하: UNet
        * 전체 오차 = image 변환오차(0%, <3%) + UNet 학습오차
        * '''최종방향: 1)설계상 속도, 2)전체과거 평균속도, 3) 최근 평균속도(10초, 30초, 60초)에 비하여 UNet이 성능이 얼마나 좋은지 확인하기''' => 5/21(금)까지
        * 추후 개선방안 : 1) feature 추가, 2) UNet tuning, 3) decoding에서 sequence(pixel) to one 모델 개발(min, avg, med, max, 중점값, LSTM 등)
        * 주영하: UNet++
        * '''2차 (1/21) – 주영하 - UNet++'''
        * Zhou, Z., Siddiquee, M. M. R., Tajbakhsh, N., & Liang, J. (2019). Unet++: Redesigning skip connections to exploit multiscale features in image segmentation. IEEE transactions on medical imaging, 39(6), 1856-1867.

Found 2 matching pages out of 185 total pages