http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
물웅덩이 탐지를 위한 YOLO 모델 성능 향상 전략 탐색
박호열(Ho-Yeol Park),배소리(So-Ri Bae),장병철(Beong-Chul Chang),차재혁(Jae-Hyuk Cha) 대한전자공학회 2023 대한전자공학회 학술대회 Vol.2023 No.6
This paper aims to explore strategies for improving the performance of the YOLO model in detecting water puddles with irregular shapes. By utilizing the preprocessed weights of the state-of-the-art YOLOv8 model, various attempts, including hyperparameter tuning, data augmentation, and annotation methods, are made to achieve outstanding performance in detecting water puddles. Finally, the study investigates the limitations of water puddle detection using the YOLO model and concludes the paper by setting directions for future research.