Dirt detection method

Abstract

The invention discloses a stain detection method. The method comprises the following steps: acquiring an image of a to-be-detected cover; performing circular target calibration on the grayscale imageand calculating a central point and a radius of the circular target image; Dividing the circular target image into a circular ring area and a central circular area, and performing polar coordinate transformation on the circular ring area image by taking the central point as an original point; carrying out adaptive binarization filtering on the image after polar coordinate transformation and the image of the central circle area to obtain stain images of the circular ring area and the central circle area respectively, and carrying out coordinate inverse transformation on the stain images under the polar coordinate system to obtain a stain image of the circular ring area under the rectangular coordinate system; and combining the stain images of the two areas to obtain a stain image of the to-be-detected cover. According to the method, the area with non-uniform illumination in the rectangular coordinate system is converted into the area with uniform illumination in the polar coordinate system through the coordinates, so that the object structure interference and the over-detection problem caused by the object structure interference are effectively avoided, and the detection precision and reliability are greatly improved.

Description

technical field [0001] The invention relates to the technical field of visual detection, in particular to a stain detection method. Background technique [0002] With the extensive development and application of computer image processing, visual image technology has been more and more widely used in the field of product defect detection, especially for the defect detection of plastic covers such as bottle caps and bucket caps; detection is mostly figure 1 The detection device performs defect detection, and uses a region segmentation algorithm to analyze the image of the cover body, and then determines whether the cover body is qualified; however, the above detection device and analysis algorithm have the following defects: [0003] 1. Existing detection devices usually only use a group of light sources for detection, and the existing detection method is to perform area segmentation and spatial filtering in the Cartesian coordinate system. However, for the special structure ...

AI Technical Summary

Problems solved by technology

[0003] 1. Existing detection devices usually only use a group of light sources for detection, and the existing detection method is to perform area segmentation and spatial filtering in the Cartesian coordinate system. However, for the special structure inside the cover, the cover area The uneven illumination from the inside to the outside causes the features of the inner cover to have undulating contours similar to the features of stains, making it impossible to achieve stain detection through the comparison of foreground and background.

[0004] 2. The existing detection method uses a multi-region segmentation algorithm. Since the edge of the cover and the boundaries of each region have very obvious grayscale changes, it is easy to cause false detection in the spatial filtering this time, and it will lead to very difficult parameter adjustments. difficulty

[0005] 3. The existing detection method needs to artificially set up multiple areas, resulting in each area needing to be debugged and set parameters separately, the workload is huge, and the detection efficiency is low

Images