Image acquisition method for acquiring all surface images of spherical material by using concave mirror

Abstract

The invention relates to a spherical material surface image acquisition method, in particular to an image acquisition method for acquiring all surface images of a spherical material by using a reflector, and belongs to the technical field of machine vision sorting. A spherical material full-surface image is obtained through one-time imaging of a single camera, and rapidness and convenience are achieved. In a set image composed of a front image (4) and three back images (5) closely arranged around the front image in a circular array mode, each back image (5) is tangent to the front image (4), the diameter of the front image (4) is the same as the size of each back image (5) in the radiation direction, and the proportional conversion processing amount is reduced. The aperture of the concavemirror selects a critical value to reduce invalid background segmentation. Aligning to the imaging background of the camera, except for the spherical material, the areas which are easy to be reflectedby the concave mirror and then enter the camera are black, and invalid background segmentation is reduced. In the collected set image, the back image is closely and circularly arrayed around the front image, so that effective pixels in each image are centralized, the area of a CCD photosensitive device is fully utilized, and the pixel utilization rate is improved.

Description

technical field [0001] The invention relates to a method for collecting surface images of spherical materials, in particular to a method for collecting images of the entire surface of spherical materials by using mirrors, and belongs to the technical category of machine vision sorting. Background technique [0002] When machine vision sorting, it is often based on the surface information of the material, and the surface information is mainly image information. When screening a large number of materials one by one, quickly obtaining all and real images of the material surface is the key to achieving online and accurate sorting. [0003] When the material is spherical or quasi-spherical, limited by the curved surface, multiple images can cover the entire material surface with multiple or multi-view imaging, but limited images theoretically have local features on the spherical material surface. A certain degree of distortion. In order to obtain the surface information of sphe...

AI Technical Summary

Problems solved by technology

Restricted by the optical characteristics of the flat mirror, there are size changes between the reflected image and the front image, especially when this technology is applied to spherical materials, the three images cannot cover the entire spherical surface, resulting in missed sampling

Compared with the plane reflector, the parabolic reflector can obtain the back image of the spherical material, and has the effect of magnifying the reflected image. However, when the magnification ratio is not controlled, the surface image processing involves complex data conversion, resulting in a large amount of calculation. long data processing time

Images