Liquid medicine bottle foreign matter detection method based on machine visual detection platform

Abstract

The invention discloses a liquid medicine bottle foreign matter detection method based on a machine visual detection platform. The method comprises the following steps: step (1), continuously acquiring N frames of sequence liquid medicine bottle images, N being a positive integer greater than or equal to one; step (2), performing mean value filtering and windowing processing on each frame of the acquired sequence liquid medicine bottle images; step (3), obtaining a detection region of interest (ROI) of each frame of the sequence liquid medicine bottle images; step (4), establishing a reference template, performing template matching based on gray scale characteristics on the liquid medicine bottle images after the filtering and the windowing processing and the reference template; and step (5), performing inter-frame differential processing on two adjacent frames of the images after the matching; performing gray scale threshold segmentation on the images after the inter-frame differential processing to obtain an Imagethreshold, i being greater than and equal to 2 and smaller than and equal to N; if no foreign matters exist in the Imagethreshold, determining that the liquid in a medicine bottle is qualified; and otherwise, determining whether the liquid in the medicine is qualified according to the characteristics of the foreign matters.

Description

technical field [0001] The invention belongs to machine vision detection technology, and relates to a method for detecting foreign objects in liquid medicine bottles based on a machine vision detection platform. Background technique [0002] Aqueous injection preparations are very common pharmaceutical preparations, and now my country's production of medical injections ranks among the top in the world. However, in the production process of aqueous injection preparations, due to the influence of the technological level, other foreign matters such as glass shavings, fibers, white spots, color blocks and crystal precipitation often inevitably appear in the aqueous injection preparations. These impurities will have a great impact on the quality of medicines. Since the mechanical platform will produce unavoidable mechanical vibrations, there will be a translation or twist of several pixels reflected in the collected image of the medicine bottle. If the image is not accurately c...

AI Technical Summary

Problems solved by technology

[0003] The present invention proposes a method for detecting foreign matter in liquid medicine bottles based on a machine vision detection platform. Calibration was performed on the moved medicine bottle, and the light interference area was not excluded, and the air bubble interference was not excluded, resulting in the problem of low detection accuracy of foreign objects and high false detection rate.

Images