Metal Surface Defect Detection Using Modified YOLO
Aiming at the problems of inefficient detection caused by traditional manual inspection and unclear features in metal surface defect detection, an improved metal surface defect detection technology based on the You Only Look Once (YOLO) model is presented. The shallow features of the 11th layer in t...
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-08-01
|
| Series: | Algorithms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4893/14/9/257 |
| _version_ | 1797520538549092352 |
|---|---|
| author | Yiming Xu Kai Zhang Li Wang |
| author_facet | Yiming Xu Kai Zhang Li Wang |
| author_sort | Yiming Xu |
| collection | DOAJ |
| description | Aiming at the problems of inefficient detection caused by traditional manual inspection and unclear features in metal surface defect detection, an improved metal surface defect detection technology based on the You Only Look Once (YOLO) model is presented. The shallow features of the 11th layer in the Darknet-53 are combined with the deep features of the neural network to generate a new scale feature layer using the basis of the network structure of YOLOv3. Its goal is to extract more features of small defects. Furthermore, then, K-Means++ is used to reduce the sensitivity to the initial cluster center when analyzing the size information of the anchor box. The optimal anchor box is selected to make the positioning more accurate. The performance of the modified metal surface defect detection technology is compared with other detection methods on the Tianchi dataset. The results show that the average detection accuracy of the modified YOLO model is 75.1%, which ia higher than that of YOLOv3. Furthermore, it also has a great detection speed advantage, compared with faster region-based convolutional neural network (Faster R-CNN) and other detection algorithms. The improved YOLO model can make the highly accurate location information of the small defect target and has strong real-time performance. |
| first_indexed | 2024-03-10T07:58:13Z |
| format | Article |
| id | doaj.art-8d49f35c77b243629115a864c1e65e37 |
| institution | Directory Open Access Journal |
| issn | 1999-4893 |
| language | English |
| last_indexed | 2024-03-10T07:58:13Z |
| publishDate | 2021-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Algorithms |
| spelling | doaj.art-8d49f35c77b243629115a864c1e65e372023-11-22T11:40:13ZengMDPI AGAlgorithms1999-48932021-08-0114925710.3390/a14090257Metal Surface Defect Detection Using Modified YOLOYiming Xu0Kai Zhang1Li Wang2College of Electrical Engineering, Nantong University, Nantong 226019, ChinaCollege of Electrical Engineering, Nantong University, Nantong 226019, ChinaCollege of Electrical Engineering, Nantong University, Nantong 226019, ChinaAiming at the problems of inefficient detection caused by traditional manual inspection and unclear features in metal surface defect detection, an improved metal surface defect detection technology based on the You Only Look Once (YOLO) model is presented. The shallow features of the 11th layer in the Darknet-53 are combined with the deep features of the neural network to generate a new scale feature layer using the basis of the network structure of YOLOv3. Its goal is to extract more features of small defects. Furthermore, then, K-Means++ is used to reduce the sensitivity to the initial cluster center when analyzing the size information of the anchor box. The optimal anchor box is selected to make the positioning more accurate. The performance of the modified metal surface defect detection technology is compared with other detection methods on the Tianchi dataset. The results show that the average detection accuracy of the modified YOLO model is 75.1%, which ia higher than that of YOLOv3. Furthermore, it also has a great detection speed advantage, compared with faster region-based convolutional neural network (Faster R-CNN) and other detection algorithms. The improved YOLO model can make the highly accurate location information of the small defect target and has strong real-time performance.https://www.mdpi.com/1999-4893/14/9/257surface defect detectionYOLOK-Means++ |
| spellingShingle | Yiming Xu Kai Zhang Li Wang Metal Surface Defect Detection Using Modified YOLO Algorithms surface defect detection YOLO K-Means++ |
| title | Metal Surface Defect Detection Using Modified YOLO |
| title_full | Metal Surface Defect Detection Using Modified YOLO |
| title_fullStr | Metal Surface Defect Detection Using Modified YOLO |
| title_full_unstemmed | Metal Surface Defect Detection Using Modified YOLO |
| title_short | Metal Surface Defect Detection Using Modified YOLO |
| title_sort | metal surface defect detection using modified yolo |
| topic | surface defect detection YOLO K-Means++ |
| url | https://www.mdpi.com/1999-4893/14/9/257 |
| work_keys_str_mv | AT yimingxu metalsurfacedefectdetectionusingmodifiedyolo AT kaizhang metalsurfacedefectdetectionusingmodifiedyolo AT liwang metalsurfacedefectdetectionusingmodifiedyolo |