Vision-based steel ball surface defect detection apparatus and detection method

Abstract

The invention relates to a vision-based steel ball surface defect detection apparatus and a vision-based steel ball surface defect detection method. According to the apparatus, a light source is arranged on a work bench; a left camera and a right camera are symmetrically arranged on the two sides of the light source; a vibration material supplying tray and a material conveying channel are arranged behind the light source; the material conveying channel penetrates a light source bowl cover and extends to the front side of the light source; a sorting mechanism and collecting boxes positioned below the sorting mechanism are arranged on the front side of the light source; a discharging terminal of the material conveying channel is inserted into a feeding terminal of the sorting mechanism; the left camera and the right camera are respectively connected to a PCI card, and the PCI card is respectively connected to a computer and a pneumatic unit; and the output end of the pneumatic unit is connected to the sorting mechanism. The method comprises the steps that: a template image is collected; the steel balls are driven to move along the material conveying channel; the left camera and the right camera respectively acquire steel ball images 3 times, and the images are compared with the template image; according to the comparison results, the PCI card controls a solenoid valve which controls the sorting mechanism, such that the detected steel ball can fall into a corresponding collecting box. According to the invention, the detection process is fully automated, and the defect detection is accurate and highly efficient.

Description

technical field [0001] The invention relates to a steel ball surface defect detection. In particular, it relates to a vision-based steel ball surface defect detection method and device. Background technique [0002] Steel balls are widely used both as bearing balls and as general-purpose rolling elements. The surface quality of the steel ball directly affects the precision and service life of the bearing. At present, domestic steel ball manufacturers mostly use light inspection in the traditional sense to detect steel ball surface defects. ) observation, and pick out defective balls manually. In the face of batch inspection requirements, the light inspection method is not only low in detection efficiency, but also cannot ensure that the ball surface is fully expanded by manual pushing, which is easy to miss or falsely detect, and at the same time causes great damage to the eyes of the operator. With the development of detection technology for steel ball surface defects, ...

AI Technical Summary

Problems solved by technology

The expansion device of this method is complicated, and the expansion is random, which cannot guarantee the complete expansion of the spherical surface; the immersion of the liquid will cause some pollution to the steel ball itself, and the collected images are pr

Images