Research on Aluminum Profile Surface Defect Detection Based on Deep Learning
Aiming at the low accuracy and efficiency of aluminum profile surface defect detection,this paper propo-ses a method of aluminum profile surface defect detection based on Yolox deep learning.Yolox-s deep learning model was adopted for training,and the existing defect data set was expanded by rotating,adjusting brightness,contrast and other methods.70％,20％and 10％of the expanded data sets were used in the training set,verifica-tion set and test set respectively for experimental verification.The results show that compared with the classical Yolov3,Yolov5 and Yolov7 algorithms,the defect detection accuracy of this method is as high as 90％,which is suitable for the aluminum profile industry.
deep learningYolox-saluminumprofile defect detection
万方数据
维普
