Research on Recognition Technology of Aluminum Profile Surface Defects Based on Deep Learning

• Published in: Materials Vol. 12; no. 10; p. 1681
• Main Authors: Wei, Ruofeng, Bi, Yunbo
• Format: Journal Article
• Language: English
• Published: Switzerland MDPI AG 23.05.2019; MDPI
• Subjects: Algorithms; Aluminum; Aluminum alloys; aluminum profile surface defects; average precision (AP); Classification; Datasets; Deep learning; Defects; Depth profiling; Inspection; Machine vision; Morphology; multiscale defect-detection network; Neural networks; Product safety; saliency maps; Steel products; Surface defects; Vision systems; deep learning; average precision (AP); multiscale defect-detection network; saliency maps; aluminum profile surface defects
• ISSN: 1996-1944; 1996-1944
• DOI: 10.3390/MA12101681

Aluminum profile surface defects can greatly affect the performance, safety, and reliability of products. Traditional human-based visual inspection has low accuracy and is time consuming, and machine vision-based methods depend on hand-crafted features that need to be carefully designed and lack robustness. To recognize the multiple types of defects with various size on aluminum profiles, a multiscale defect-detection network based on deep learning is proposed. Then, the network is trained and evaluated using aluminum profile surface defects images. Results show 84.6%, 48.5%, 96.9%, 97.9%, 96.9%, 42.5%, 47.2%, 100%, 100%, and 43.3% average precision (AP) for the 10 defect categories, respectively, with a mean AP of 75.8%, which illustrate the effectiveness of the network in aluminum profile surface defects detection. In addition, saliency maps also show the feasibility of the proposed network.