用於不平衡口罩數據集的三階段遞迴式訓練技術

Three-stage recursive training technique for imbalanced face mask datasets

施維宣(Wei-Hsuan Shih)

指導教授 ： 蔡奇謚

淡江大學/工學院/電機工程學系碩士在職專班/碩士(2024年)

https://doi.org/10.6846/tku202400015

摘要

本篇摘要未授權

關鍵字

不平衡數據 ； 遞迴式學習 ； 口罩偵測

並列摘要

This abstract would not be displayed due to no authorization.

並列關鍵字

Imbalanced data ； Recursive learning ； Face mask detection

參考文獻

[1] YOLOv6 v3.0: A Full-Scale Reloading. Available online: https://github.com/meituan/YOLOv6.

[2] YOLOv7: Trainable Bag-of-Freebies Sets New State-of-the-Art for Real-Time Object Detectors. Available online: https://github.com/WongKinYiu/yolov7.

[3] YOLO by Ultralytics (Version 8.0.0). Available online: https://github.com/ultralytics/ultralytics.

[4] YOLOX: Exceeding YOLO Series in 2021. Available online: https://github.com/Megvii-BaseDetection/YOLOX.

[5] Chawla, N. V., Bowyer, K. W., Hall, L. O. Kegelmeyer, W. P. "SMOTE: Synthetic Minority Over-sampling Technique." Journal of Artificial Intelligence Research, vol. 16, 2002, pp. 321-357, doi:10.1613/jair.953