แจ้งเอกสารไม่ครบถ้วน, ไม่ตรงกับชื่อเรื่อง หรือมีข้อผิดพลาดเกี่ยวกับเอกสาร ติดต่อที่นี่ ==>
หากไม่มีอีเมลผู้รับให้กรอก thailis-noc@uni.net.th
หากไม่มีอีเมลผู้รับให้กรอก thailis-noc@uni.net.th
Park, PooGyeon. Multi-output object detection network for automated categorization and grading of steelmaking defects. (). King Mongkut's University of Technology North Bangkok. Central Library. : , 2026.
Multi-output object detection network for automated categorization and grading of steelmaking defects
Name: Ryu, Seung Hyun
Organization :
POSTECH. Dept. of Electrical Engineering
Email :
yu4546@postech.ac.kr
Name: Park, Chan
Organization :
POSTECH. Dept. of Electrical Engineering
Email :
parkchan@postech.ac.kr
Name: Kim, Kyungsoo
Organization :
POSTECH. Dept. of Electrical Engineering
Email :
kyungsoo@postech.ac.kr
Name: Park, PooGyeon
Organization :
POSTECH. Dept. of Electrical Engineering
Email :
ppg@postech.ac.kr
keyword:
Object detection (Computer vision)
LCSH:
Quality control
-- Automation
Abstract:
This study introduces an integrated deep learning architecture designed to perform concurrent categorization and severity assessment of steelmaking defects. In steel manufacturing, achieving high classification accuracy is often compromised by the visual similarity between critical and non-critical defects, which directly impacts quality control reliability. By incorporating a grading branch in parallel with the classification layer within a multi-output object detection framework, the model facilitates a unified diagnostic process. To address the inherent complexities of surface anomalies, specialized data augmentation and bounding box expansion strategies are introduced. By refining the spatial representation of defects, these methods enable the network to focus on subtle discriminative features that are crucial for accurate class and grade determination. Experimental results using industrial steel surface datasets demonstrate that the proposed unified approach significantly enhances the precision of both defect categorization and quality grading.
King Mongkut's University of Technology North Bangkok. Central Library
Address:
BANGKOK
Email:
library@kmutnb.ac.th
Created:
2026
Modified:
2026-08-04
Issued:
2026-08-04
บทความ/Article
application/pdf
BibliograpyCitation :
In ECTI Association and Burapha University. Faculty of Engineering. 2026 23rd International Conference on Electrical Engineering/Electronics, Computer, Telecommunications and Information Technology (ECTI-CON 2026) (pp.60-64). Piscataway, NJ : IEEE
eng
©copyrights King Mongkut's University of Technology North Bangkok
RightsAccess:
ใช้เวลา
0.045536 วินาที
Ryu, Seung Hyun
| Title | Contributor | Type |
|---|---|---|
| Multi-output object detection network for automated categorization and grading of steelmaking defects
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Ryu, Seung Hyun;Park, Chan;Kim, Kyungsoo;Park, PooGyeon | บทความ/Article | |
| An empirical study of action and observation spaces for imitation learning in contact-rich pushing
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Kim, KyungSoo;Lee, Hye Jin;Ryu, Seung Hyun;Park, PooGyeon | บทความ/Article |
Park, Chan
Kim, Kyungsoo
| Title | Contributor | Type |
|---|---|---|
| Multi-output object detection network for automated categorization and grading of steelmaking defects
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Ryu, Seung Hyun;Park, Chan;Kim, Kyungsoo;Park, PooGyeon | บทความ/Article | |
| An empirical study of action and observation spaces for imitation learning in contact-rich pushing
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Kim, KyungSoo;Lee, Hye Jin;Ryu, Seung Hyun;Park, PooGyeon | บทความ/Article |
Park, PooGyeon
| Title | Contributor | Type |
|---|---|---|
| Admissibility criterion for T-S fuzzy descriptor systems with time-varying delays using free-matrix-based integral inequality
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Lee, Hae Seong;Park, Chan;Park, PooGyeon | บทความ/Article | |
| An active noise control algorithm using diffusion adaptive filter robust to impulsive noise and input noise in sparse systems
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Park, Chan;Lee, Hae Seong;Park, PooGyeon | บทความ/Article | |
| Multi-output object detection network for automated categorization and grading of steelmaking defects
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Ryu, Seung Hyun;Park, Chan;Kim, Kyungsoo;Park, PooGyeon | บทความ/Article | |
| An empirical study of action and observation spaces for imitation learning in contact-rich pushing
มหาวิทยาลัยเทคโนโลยีพระจอมเกล้าพระนครเหนือ Kim, KyungSoo;Lee, Hye Jin;Ryu, Seung Hyun;Park, PooGyeon | บทความ/Article |