Research on Intelligent Detection Method of Forging Magnetic Particle Flaw Detection Based on YOLOv4

Aiming at the problems of low efficiency and low accuracy of defect detection in forging manufacturing enterprises, an intelligent detection method of magnetic particle flaw detection based on YOLOv4 is proposed. This paper designs the implementation scheme of intelligent detection system and the evaluation experiment of detection effect. Firstly, the image acquisition platform of fluorescent magnetic particle flaw detection is built, and the defect sample data set is constructed. The results of comparative experiments show that the mAP of YOLOv4 on the effective data set is 91.73%, the F1 score is 0.91, and the floating-point operand is 3.8B. Compared with other deep learning models, this method has advantages in speed and accuracy, and can meet the needs of relevant production enterprises.