Redirigiendo al acceso original de articulo en 24 segundos...
Inicio  /  Algorithms  /  Vol: 14 Par: 9 (2021)  /  Artículo
ARTÍCULO
TITULO

Metal Surface Defect Detection Using Modified YOLO

Yiming Xu    
Kai Zhang and Li Wang    

Resumen

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.

Palabras claves

 Artículos similares

       
 
Xuxing Huang, Xuefeng Li, Hequn Li, Shanda Duan, Yihao Yang, Han Du and Wuning Xiao    
The goaf treatment of underground metal mines is an important link in mining, and it is particularly important to master the laws of overlying rock strata and surface movement of goaf. In this paper, Persistent Scatterer Interferometric Synthetic Apertur... ver más
Revista: Applied Sciences

 
Zhili Wang, Lan Liang, Ning Li, Shuang Wu, Zhanjun Cheng, Beibei Yan and Guanyi Chen    
Graphite carbon nitride (g-C3N4) has been employed as an emerging metal-free catalyst in heterogeneous catalysis. However, the catalyst has a poor activation property for peroxymonosulfate (PMS). In this study, Bi-Fe oxide co-doped g-C3N4 (Bi@Fe/CN) was ... ver más
Revista: Applied Sciences

 
L. P. Cahalan, M. B. Williams, L. N. Brewer, M. M. McDonnell, M. R. Kelly, A. D. Lalonde, P. G. Allison and J. B. Jordon    
Large-scale metal additive manufacturing (AM) provides a unique solution to rapidly develop prototype components with net-shape or near-net shape geometries. Specifically, additive friction stir deposition (AFSD) is a solid-state method for large-scale m... ver más
Revista: Applied Sciences

 
Xinyue Yan, Xin Chen, Wenyan Zheng, Guilin Zhang and Aiguo Dong    
Fifty-one surface sediment samples from Dongshan Bay, China, were analyzed for heavy metals to evaluate their distribution, pollution status, and controlling factors. The enrichment factor is suggestive of the potential pollution status, ranging from min... ver más
Revista: Water

 
Aminur Rahman    
This study explores the potential of modified shrimp-based chitosan (MSC) as an innovative adsorbent for eliminating heavy metals (HMs) from contaminated water sources. The modifications encompassed various chemical treatments, surface functionalization,... ver más
Revista: Water