Contact lens edge defect detection method

Abstract

The invention discloses a contact lens edge defect detection method. The detection method comprises the following steps: collecting a two-dimensional image of a contact lens; preprocessing the two-dimensional image; performing transverse and longitudinal projection on the preprocessed two-dimensional image to obtain a one-dimensional signal; performing second-order derivation on the one-dimensional signal to obtain edge geometric parameters of the contact lens; performing space polar coordinate transformation by utilizing the edge geometric parameters; extracting an edge line according to thepolar coordinates; and extracting characteristic parameters of the extracted edge lines, and if the characteristic parameters exceed a preset threshold value, judging that the edges of the contact lenses have defects. The circle center and the radius of the corneal contact lens are obtained in a short time through derivation after transverse and longitudinal pixels in the image are accumulated, the contour of the circular image is extracted on the basis, tiny defect information can be prevented from being lost, and compared with the prior art, the calculation speed is high, and defect judgmentis accurate.

Description

technical field [0001] The invention belongs to the field of biomedical engineering and relates to a detection method for edge defects of contact lenses. Background technique [0002] Contact lens is a kind of soft lens widely used for vision correction. Its main manufacturing methods are turning and injection molding. Due to the toughness of the material, defects such as injection cracks, bubbles, and mold brightness are prone to occur during processing. , causing discomfort to the wearer, inflammation of the corners of the eyes and even corneal damage. The image acquisition system based on machine vision can obtain the shape image of the contact lens, and automatic defect detection can be realized through image enhancement, edge extraction and feature analysis. [0003] The edge shape of the contact lens is circular, and circle detection is required to extract the edge. The existing circle detection methods include Hough transform, random Hough transform, least square met...

AI Technical Summary

Problems solved by technology

[0003]The edge shape of the contact lens is circular, and the circle detection is required to extract the edge. The existing circle detection methods include Hough transform, random Hough transform, least squares Multiplication circle fitting, etc., among which Hough transform is a method that uses parameter transformation to carry out information statistics to obtain the coordinates and radius of the circle center. This method has a large time complexity and space complexity, and the calculation speed of this method is slow. The extracted The contour is easy to lose tiny edge defects, and the calculation speed and defect detection accuracy are insufficient; the random Hough transformation improves the classic Hough transformation, but this method is not suitable for ring detection; the least square method circle fitting is relatively , the speed is faster, but the image processing is difficult due to the influence of the shape, pattern, front and back of the contact lens, etc.

In addition, the edge defects of contact lenses include mold edges, broken edges, cracked edges, turning edges, deformation, etc. Existing detection methods only extract the distance from the edge to the center of the circle for defect analysis, which is likely to cause missed and false detections

Images