Bearing Defect Detection with Unsupervised Neural Networks

Algorithm 1

A normalized samples method to obtain the effective part of inner diameter sample from the raw bearing samples.

	Input: inner diameter sample with 1280 × 1024 pixels.
	Output: normalized samples of inner diameter sample with 760 × 760 pixels.
(1)	Morphological denoising: the original image is corroded and expanded, and the template is a 5 ∗ 5 rectangular morphological structural element;
(2)	Binarize the original image, take the maximum gray value and minimum gray value of the inner diameter area as the threshold, set the image greater than the maximum threshold and less than the minimum threshold to 255, and the inner diameter area becomes 0;
(3)	Search the inner edge contour, and then fit the inner edge with an ellipse;
(4)	Use the ellipse fitted in step 3 to remove the extra part of the image;
(5)	Search the four points at the top, bottom, left, and right of the inner edge;
(6)	Map the above ellipse to a circle. Take the four points of the top, bottom, left, and right of the circle with the center of the image as the center and the radius of 290 as the target points to establish a projection transformation mapping matrix, and then use this transformation matrix to transform the image in step 4;
(7)	Search for the outer edge contour, fit the ellipse, and cut off the outside of the ellipse;
(8)	Search the area of the inner diameter, and cut off the outer part of the inner diameter area.