Detectors++: The Robust Baseline for a Defect Detection System

Fityanul Akhyar, Chih Yang Lin, Gugan S. Kathiresan, Bharath Surianarayanan, Chao Yung Hsu

研究成果: 書貢獻/報告類型會議論文篇章同行評審

3 引文 斯高帕斯(Scopus)

摘要

Focusing on the task of steel surface defect localization, this study employs the latest state-of-the-art RCNN family, Cascade RCNN, on top of the current FPN model called DetectoRS-ResNeXt. The baseline was tested individually using Side Aware Boundary Localization (SABL) plus pixel domain augmentation to obtain the precision of predictions. Trained on a well-known real-world dataset, Severstal, our proposal achieves a mAP of 82.5% which offers the potential to serve as a high-quality defect detection baseline.

原文???core.languages.en_GB???
主出版物標題2021 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2021
發行者Institute of Electrical and Electronics Engineers Inc.
ISBN(電子)9781665433280
DOIs
出版狀態已出版 - 2021
事件8th IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2021 - Penghu, Taiwan
持續時間: 15 9月 202117 9月 2021

出版系列

名字2021 IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2021

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???8th IEEE International Conference on Consumer Electronics-Taiwan, ICCE-TW 2021
國家/地區Taiwan
城市Penghu
期間15/09/2117/09/21

指紋

深入研究「Detectors++: The Robust Baseline for a Defect Detection System」主題。共同形成了獨特的指紋。

引用此