Inductor quality evaluation method and system based on artificial intelligence, and storage medium

The invention discloses an artificial intelligence-based inductor quality evaluation method and system and a storage medium, and the method comprises the steps: obtaining the production step image data of an inductor device through an image processing technology in a preset production period, and recognizing defect information through a CNN model; then, tracking and analyzing the defects at the same position, and grouping the defects by utilizing a spectral clustering algorithm; and after the period is finished, marking the unqualified inductor devices, and screening out the associated defect group. And in the next period, combining the new defect information with the associated defect data set to carry out defect influence analysis and production quality evaluation so as to improve the product quality. By means of the inductor production defect tracking method and device, defect tracking of inductor production can be effectively achieved, accurate classification and judgment can be conducted on defects, info