Machine vision system for detecting bottom of large empty infusion bottle and operation method of machine vision system

Abstract

The invention discloses a machine vision system for detecting the bottom of a large empty infusion bottle and an operation method of the machine vision system. The machine vision system comprises an imaging unit, a detection sample and a light source assembly; the detection sample is located at the top of the light source assembly; the imaging unit is located at the top of the detection sample, and comprises a camera, an imaging target surface and an imaging lens; the imaging target surface is installed in the camera, the imaging lens is installed at the bottom of the camera, and a light source plate is fixedly sleeved with a fixed lantern ring. According to the machine vision system and the operation method thereof, constraint conditions such as the working distance, the magnification times and the detection view field in vision imaging are constructed according to shape information of the empty large infusion bottle and the range and precision requirements of the bottle bottom to be detected, a machine vision model for bottle bottom detection is established, clear imaging of the bottle bottom can be realized, and interference of inherent accessories of the bottle body on imaging of the bottle bottom in a traditional detection mode is overcome.

Description

technical field [0001] The invention relates to the field of bottle bottom detection in empty liquid bottles, in particular to a machine vision system and an operation method for detecting the bottom of large infusion liquid bottles. Background technique [0002] The possible foreign matter in the large infusion needs to be inspected bottle by bottle, including the inspection of empty bottles before filling and the inspection of liquid bottles after filling. For the inspection of empty bottles before filling, it mainly includes the inspection of the bottle body and the bottom of the bottle. For the detection of the bottle body, the visual model and detection method are relatively fixed, that is, the bottle body is between the backlight and the imaging unit, and the information of foreign objects that may exist in the bottle body is collected in the imaging unit, but for the detection of the bottom of the bottle, Due to the special shape of the large infusion bottle and the s...

AI Technical Summary

Problems solved by technology

Among them, the imaging unit is located above the bottle mouth and passes through the bottle mouth to realize the imaging of the bottom of the bottle. However, this method simply captures the image of the bottom of the bottle through the imaging unit, but does not take into account the constraints of detection accuracy and the shape of the bottle mouth.

This method is often used for bottle bottom detection with low detection accuracy or small detection field of view.

In addition, for the second method, since the bottom of the infusion bottle has a special structure such as a pull ring, if the imaging unit is directly aimed at the bottom of the bottle for imaging from the bottom of the bottle, interference information such as the pull ring will be imaged in the image, resulting in detection due to Ineffective due to too much noise

Images