Circular contour detection method under condition that circular gold surface of flexible IC substrate is missing

Abstract

The invention discloses a circular contour detection method under the condition that a circular gold surface of a flexible IC substrate is missing. The circular contour detection method comprises thefollowing steps: acquiring an image of the flexible IC substrate; converting the image into an HSV color space, and extracting an H-channel image as a processing object; preprocessing the image; constructing an edge type geometric activity contour model; and inputting the image into an edge type geometric active contour model, and obtaining an optimal contour by minimizing an energy functional ofa closed curve. According to the invention, a single-channel image in an HSV color space is used for processing; an edge type geometric activity contour model is constructed; according to the method,the contour detection problem is converted into the curve approximation problem, the contour detection of the round gold surface missing area of the flexible IC substrate is realized by designing theedge type geometric movable contour model, and the method can be used for solving the appearance defect detection problems of poor Mark points, poor SMT holes and the like of the flexible IC substrate.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a circular contour detection method in the case of a missing circular gold surface of a flexible IC substrate. Background technique [0002] The circular contour detection method of flexible IC substrates generally uses first-order or second-order edge detection operators, including Sobel operator, Canny operator, Laplacian operator and other methods. For the complete image of the gold surface, this method can accurately detect the circular contour. However, when the circular gold surface is missing, the contrast between the region of interest and the background decreases, the gray level difference becomes smaller, and the contour detected by the edge operator is discontinuous. In order to obtain a complete contour, Hough circle transform is usually used to simulate For contours with low completeness, the fitting result is easy to be far from the real contour. For contours with hi...

AI Technical Summary

Problems solved by technology

However, when the circular gold surface is missing, the contrast between the region of interest and the background decreases, the gray level difference becomes smaller, and the contour detected by the edge operator is discontinuous. In order to obtain a complete contour, Hough circle transform is usually used to simulate For contours with low completeness, the fitting result is easy to be far from the real contour. For contours with high completeness, although the fitting result is close to the real contour, the result of Hough circle transformation is a theoretical circle, so The result is difficult to fit the true value

Images