In the modern packaging industry automation production, involves a variety of inspection, measurement, such as beverage bottle cap printing quality inspection, product packaging bar code and character recognition. The common characteristics of such applications are continuous mass production and very high appearance quality requirements. Usually this is highly repetitive tasks can only rely on manual inspection to complete, we often see behind the modern assembly line in some factory hundreds or even more than thousand testing workers to perform this procedure, the increase for the factory huge artificial cost and management cost at the same time, still can't guarantee a 100% inspection qualified rate (i.e., "zero defect").

Today's competition between companies does not allow even a 0.1 percent defect. Sometimes, such as accurate and rapid measurement of small size, shape matching, color identification, etc., the human eye can not be carried out continuously and stably, and other physical quantity sensors are difficult to be used. At this time, people began to consider the speed of the computer, reliability, repeatability of the results, thus introducing the robot vision technology.

As we all know, food and drugs are related to human life and health. If the quality of drugs threatens human life, it will be a great disaster. Therefore, the drug manufacturers, especially the world famous big factories, have paid great attention to the whole production process and even the packaging of the back of the drug. In the production and packaging process of food and drugs, it can play a powerful role in the blister packaging of drugs, liquid filling, or the gland, labeling, coding in the back section, as well as the final detection of the box.

I. Machine vision lack of medicine or lack of bottle detection

Due to the strict regulations of the pharmaceutical industry, the quality of pharmaceutical packaging is becoming more and more demanding. When the drug particles are packed into the blister, the manufacturer must ensure that all the drug particles in the blister are intact. Alternatively, when the drug leaves the factory, a number of bottles of medicine are usually packed in a larger package, and the manufacturer must ensure that there is no shortage of bottles in each package to avoid the impact on the reputation of the drug manufacturer. Solution: the use of machine vision method, the object can be detected quickly and accurately is intact, by setting the image sensor, object images after the packaging information, through the area of the preset parameters on the grain or the medicine bottle for testing, so that the damaged medicine granule or lack of bottle packaging will be detected, right through properly.

Two, machine vision bottle mouth damage detection

Liquid medicine bottle, after canned, to determine whether there is damage to the bottle mouth, which is related to whether the liquid medicine will be mixed with glass debris. After the image sensor is installed in the liquid medicine canning process, the image matching tool is used to determine whether the bottle mouth is damaged. Before detection, the image sensor records the normal characteristics of the bottle mouth. When the canned medicine bottle passes in front of the sensor lens, the sensor will capture the current characteristics of the bottle mouth and compare them with the memorized characteristics of the original bottle mouth to see if they are consistent. If they are different, the sensor will send a signal to eliminate the bottle by the eliminating mechanism. The user can set the similarity degree according to the characteristics of the bottle mouth through the visual software. Suppose it is set as 90%, that is to say, when the similarity between the characteristics of the detected bottle mouth and the features remembered by the sensor reaches 90% or more, the sensor will determine that the bottle mouth is intact. With this test, all bottles with broken edges can be removed.

3. Machine vision potting quality inspection

Another concern on the pharmaceutical canning line is whether the lid is pressed in place after the crimping. Is the amount of liquid filling sufficient? To ensure that the bottle closed intact, ensure the vacuum degree of the bottle; Also make sure the dosage is correct. Solution: After the image sensor is installed in the gland process, the linear tool is used to measure the change of the cap and liquid level in the Y-axis direction to determine whether the cap is installed in place and whether the dosage is correct. By measuring the gap between the cap and the mouth of the bottle to determine whether the cap is installed in place; The liquid level is judged by measuring the distance between the liquid level and the mouth of the bottle. They are relative measurements, so they are not affected by the faint beating of the bottle on the conveyor belt. After this test, it can ensure that the bottle cap is not installed in place and the medicine liquid is not enough bottles are removed.

The above is only listed in the food and drug industry in part of the application, promote machine vision detection, the first thing to make food and drug enterprises more understanding of machine vision. Where to install, how to install, how to achieve the best effect, what specific benefits can bring, all these need to be constantly popularized and publicized. Practitioners in the printing and packaging industry should put forward their own specific needs, learn and master the concept and operation of machine vision inspection; The practitioners of machine vision industry should understand the printing process, production process, operation mode and quality standards. Only the close combination of the two can truly conform to the national conditions of China's automatic visual inspection system, in order to achieve the purpose of improving quality, improving efficiency and reducing costs.