Detection method for image defect of receiver product and detection light source thereof

Abstract

The invention relates to the technical field of image defect detection, in particular to a detection method for an image defect of a receiver product and a detection light source thereof. The detection method for the image defect of the receiver product is to detect the image defect of the receiver product by using the detection light source, and the detection method comprises the following stepsof (1) debugging the light source; (2) setting light source parameters; (3), collecting images; (4), image processing; and (5), dividing a warehouse for the product. In the invention, the collected image is subjected to image processing by a photometric stereo vision algorithm, so that the gentle slope pit defect is detected, and the image forming mode displays the height information, so the dirtydefect is not presented in the image, and the pit defect rather than the dirty defect can be detected only; the influence of the dirty defect is avoided; the pit defect of the receiver product can beautomatically detected in batches, the manual inspection process is eliminated, and the detection efficiency is improved and the detection cost is saved.

Description

technical field [0001] The invention relates to the technical field of imaging defect detection, in particular to a method for detecting imaging defects of receiver products and a detection light source thereof. Background technique [0002] Photometric stereo vision technology analyzes the three-dimensional shape information of objects through two-dimensional surface texture. This technology requires at least three light sources with different angles to image the same object, and the angle of the light source must be fixed and unchangeable. Although photometric stereo vision can perform 3D surface reconstruction in the form of relative height maps, it is not suitable for high-precision absolute 3D measurement or reconstruction applications. This can cause errors if the light source does not have parallel rays with a uniform intensity distribution or if the surface of the object does not have Lambertian reflection properties. Photometric stereo vision assumes parallel ligh...

AI Technical Summary

Problems solved by technology

[0004] The light source in the prior art is an ordinary coaxial light source, which cannot show gentle slope-shaped pits on the image, and the dirt defects and pit defects captured in the same form cannot be distinguished

Due to the lack of sharpness of the gentle slope pit, the edge of the pit is not obvious, and the ordinary coaxial light source cannot show the edge of the pit; the edge of the sharp pit is obvious, which can be shown by the ordinary coaxial light source, and the pit part is more than flat The area is dark, but the dirt also appears darker than the flat area under the same light source, making the dirt defect and the pit defect indistinguishable

Images