Abstract
Steel is the most basic raw material in China’s industrial production, which plays a great role in promoting China’s industrialization process. Therefore, it is of great significance to detect defective steel and the surface quality of steel. In order to further improve the detection accuracy of steel surface defects, this paper proposes a steel surface defect detection algorithm based on global context block. In this paper, a global context module is introduced based on the UNet++ network model to achieve accurate segmentation and classification of complex steel surface defects. The results show that the improved UNet++ network model achieves a dice coefficient of 94.67% on the steel surface defect dataset provided by the Kaggle competition platform. Compared with semantic segmentation models such as UNet, LinkNet, and UNet++, the segmentation effect is more accurate. Therefore, the deep learning model based on the improved UNet++ can learn more semantic features from industrial steel images, so as to obtain more accurate steel defect information. This method can be a big help for real-world applications like defect detection in industrial images.
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
