PROBLEM TO BE SOLVED: To detect a foreign substance more easily and with very high accuracy. An image data relating to a printed substrate without foreign matter for a neural network having a coding unit that extracts a feature amount from input image data and a decoding unit that reconstructs the image data from the feature amount. Using the AI model 100 generated by training only as training data, the image data obtained by the camera 32D is input to the AI model 100 as the original image data, and the reconstructed image data is acquired as the reconstructed image data. .. The original image data and the reconstructed image data are compared, and the presence or absence of foreign matter on the printed circuit board is determined based on the comparison result. Since it is not necessary to prepare a master substrate for comparison, foreign matter inspection becomes easy. Further, since the shapes and appearances of the printed circuit boards are almost the same in both image data to be compared, foreign matter can be detected with great accuracy. [Selection diagram] Fig. 4
展开▼