Method and device for automatically detecting surface defects of spherules based on machine vision

Abstract

The invention discloses a method and a device for automatically detecting surface defects of spherules based on machine vision. The device comprises an automatic spherule distribution device, an image pickup device, an automatic blanking device and an automatic spherule rolling device, wherein the automatic spherule rolling device rotates automatically and circularly among the automatic spherule distribution device, the image pickup device and the automatic blanking device through a delivery mechanism; and the delivery mechanism is connected with a PLC which is in signal connection with a computer. The detection method comprises the following steps of: automatic classification of the sphere dimension, automatic charging and automatic array distribution of the spherules, acquisition of ever changing spherical surface images through a camera, array positioning of spherules with defects, elimination of the spherules with the defects through an eliminator, blanking, and the like. The invention drives spheres to be subjected to automatic array distribution in batches and simultaneously be subjected to automatic multidirectional rolling, makes the spherical surfaces in the whole array capable of being imaged and recognized simultaneously, does not need to take kerosene as a working medium, and is suitable for detecting spherules made of any material.

Description

technical field [0001] The invention relates to automatic detection of surface defects of spheres, in particular to a method and equipment for automatic detection of surface defects of small spheres based on machine vision. Background technique [0002] Ball bearings are widely used general-purpose mechanical products. The balls are spheres, which are in great demand, and their surface quality directly affects the quality of the bearing. my country's bearing ball manufacturers basically use manual visual inspection to detect the surface quality of the sphere, that is, manually move the sphere in the flat shovel and observe its surface with eyes (or with the help of a low-magnification magnifying glass) under the astigmatism light, and find that Defective steel balls are manually picked out. People's eyes are prone to fatigue when observing under the light, especially the highly reflective surface of the polished steel ball irritates the eyes, and the operator cannot ensure t...

AI Technical Summary

Problems solved by technology

Since the steel balls are fed one by one in this method, only one steel ball can be fed at a time, and the steel balls to be tested only occupy a diagonal position in a single row on the collected image, which limits the detection efficiency of this method and equipment, and is not suitable for detection Small steel balls (such as less than 3mm in diameter); this method needs to use kerosene as the working medium, and there is a certain degree of fouling on the steel ball itself, and the follow-up procedure for cleaning the surface of the steel ball requires frequent replacement of the kerosene medium, increasing the process steps and costs

Images