An Arc Fitting Method Based on the Snake Model and Iterative Polarity Conversion Regression

Abstract

The invention discloses an arc fitting method based on the combination of the Snake model and iterative polarity conversion regression, which is used for locating the R-angle center and solving the R-angle radius in the shaft workpiece image, and belongs to the research field of machine vision precision measurement technology. Perform image preprocessing on the area of ​​interest of the frame-selected polygonal image to eliminate isolated noise points in the original image and improve the contrast of the image; perform edge extraction through the Canny operator to achieve initial edge positioning; use the Snake model to filter out larger The false edge points generated by the burr, and interpolation edge points in the sparse area, that is, the Snake optimization of the original edge; use the Zernike orthogonal moments to fine-position the edge to obtain accurate sub-pixel-level edges; through the iterative polarity conversion regression method Perform circle fitting on the sub-pixel edge points to obtain the precise center position and radius of the R angle. The detection method solves the problem of reducing the accuracy of R-angle detection of shaft workpieces due to the interference of attachments and the reliability of short arc fitting.

Description

Technical field [0001] The present invention relates to a measuring technique of machine visual and a shaft workpiece measuring system, which specifically relates to a circular arc fitting method based on a Snake model and iterative polarity transition regression, which belongs to the field of machine visual precision measurement technology. Background technique [0002] The axial workpiece is extremely common in the industrial and very important rotating workpiece. The integrated shape error accuracy not only directly affects the movement of the machine, but also has an important impact on reducing energy consumption and environmental pollution. The R angle is the key and difficult points in the axial workpiece error detection. The reason is that the R angle of the axial workpiece is a surface, even if the telecentric lens and the parallel backlight, it is not possible to completely parallel every point on the surface. On the CCD photosensitive chip of the industrial camera, the...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to design an arc fitting method combining the Snake model and iterative polarity conversion regression for the interference of contour attachments on the R corner detection in visual measurement and the instability of short arc fitting, specifically It refers to a method that collects images of shaft workpieces through synchronous movement of industrial line array cameras and parallel backlights, and accurately locates the R-angle center and solves the R-angle radius through image processing technology, which can reliably and accurately extract the R-angle center in the region of interest. method to measure some dimensional parameters of shaft workpieces

Images