The invention discloses a printed circuit board surface defect positioning and identification method. The method comprises the following steps: firstly, collecting a to-be-detected image and a template image of a printed circuit board; secondly, using a first convolutional neural network to obtain a feature point prediction distribution map, and screening to obtain an optimal feature point; then,calculating a description vector of the optimal feature point by using a second convolutional neural network, and matching the optimal feature points of the to-be-detected image and the template image; then, calculating an affine transformation matrix according to the matching points, and projecting the image to be measured onto the template image; thirdly, calculating a power spectrum of the projected to-be-detected image and the template image, obtaining an abnormal frequency component of the to-be-detected image according to the power spectrum difference, and obtaining a suspected defect area through inverse Fourier transform; and finally, identifying and classifying the suspected defect area by using a third convolutional neural network. According to the method, the surface defects ofthe printed circuit board can be accurately positioned and identified, the quality of the circuit board is guaranteed, and the method has certain robustness to environmental noise.