Abstract
In order to solve the problem of identifying and locating the stains of steel plates before laser cleaning, a real-time monitoring system for laser cleaning of steel plate based on deep learning is proposed. First of all, this paper uses K-means to cluster the points on the image, then binarizes the image, segments the steel plate image from the background. Secondly, it uses YOLOv4 for steel plate stain detection, uses CSPDreknet53 to perform feature extraction on steel plates, performs stain detection through three different feature layers, and finally outputs stain category and coordinate information. 5000 datasets were obtained through the laser cleaning platform for model training, the division ratio of training set, validation set and test set was 8:1:1. The mAP of this method for stain detection on steel plate images is 96.37%. Compared with SSD, Faster-RCNN, and YOLOv3, the accuracy rate is better, which can meet the real-time monitoring of laser cleaning.
Sign-in/Register to access full text options 
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.
