What is machine vision?
Machine vision has developed over the last 30 years, and is today used in nearly all existing business areas, and in many different applications. Machine vision is often mentioned as a quality inspection system, but it is far more than that.
Basically, machine vision is used to analyze information hidden in pictures, and there is much more information than just to check for quality issues. Often is this information overseen and not used as it could be.
Off cause the quality inspection is a very important part of how you can increase your earnings with a machine vision system, but there are other areas – often not noticed – that potential could increase your earnings with even more value than the quality aspect.
In the following, we will try to touch different aspects of where machine vision can contribute.
The aspects are:
- Quality
- Resources
- Control
- Knowledge
- Image
As already mentioned, quality inspection is an important area where you can increase your earnings with machine vision.
- Gain new costumers – more turnover
- Competitive edge
- Higher quality – higher price
- Different qualities at different prices
- Less false rejects
A quality control system can be the way to get new customers in new areas, where at stable and controlled quality is necessary to fulfill government rules, ex. FDA in US, if you are going to supply medical devices or medicine.
A competitive edge, that makes you able to supply a unique high and stable quality, better than your competitors, with a benefit for your customers. This is often seen in the sub supplier business, where a constant high quality minimizes the errors or defects in the following production. Detection of the errors or defects at the beginning of the supply chain is the most cost-effective way for your customers to secure their earnings. Machine vision is used to secure that only parts that fulfil the customers’ demands will leave the site.
Often the customers are willing to pay a premium for a quality that minimizes their internal quality cost. They can reduce their time and resources spent on handling defects or repair of finished goods.
It is also often seen that machine vision makes it possible to have control over different qualities of a product. The products can be sold at different prices. High price for the higher quality and lower price for second quality – before it was controlled, they had to deliver at and average price, because of variation in the delivered quality, they couldn’t get the premium price.
With human quality inspection, it is sometimes difficult to keep the same quality standard from morning until night, and from person to person. In some cases, the defects are very difficult to see, and where defects are not allowed to pass, the number of false defects (goods that are ok, but seen as defect by the human eye) will increase. Here machine vision can help to secure the quality and decrease the false rejects – which will increase your earnings.
Today machine vision combined with machine learning (deep learning by neural network) it is possible to detect and classify aesthetic quality demands – where the variance from the human inspection is even higher than normal quality inspection for defects.
Around the world there are many resources used in quality control. Many employees are doing the inspection manually with the above-mentioned disadvantages. It is difficult to keep the same standard over time and from employee to employee. A machine vision system works day after day with the same quality standard and with a speed that makes it possible for you to save resources and at the same time with a more stable quality and less false rejects. This is a direct increase in your earnings.
An indirect increasing in your earnings, will probably be seen in your decreasing cost of handling customer complaints or goods that have been rejected by you customers. This can be an overseen but high cost.
With a machine vision system, you will not only be able to see quality issues, but you are able to gather information – a lot of information – about your production, process and your products. This information can be used in several ways. In the following we will mention some of the very interesting areas where the data can be used, and will help you increase your earnings.
- Control of machines
- Statistics
- Knowledge about your production equipment
- Preventive maintenance
In several applications machine vision is used to check quality and through a feedback loop to control the same machine to keep or adjust it to the wanted quality. This could be controlling a laser printing on parts or cutting parts. If the laser is not performing as wanted, the machine vision can ‘see’ what is wrong and correct the laser through feedback. This gives less rejected goods and will increase your earnings.
Machine vision can be used to detect objects, that has to be cut out or altered, and give geometric information (ex. X,Y,Z) to following processing machines. Alternative employees have to mark and give information to the following machines – time consuming and expensive.
There are a lot of information in the statistics that are possible to make, when you are using a machine vision system. Statistic material can tell you about variation over time and give you a tool to control the process. Smaller variations can be seen earlier, and therefore you will be able to adjust the process before it is visible in defective products. The preventive adjustment will save you for defect cost and indirect increase your earnings.
It some situations it is possible to measure the real accuracy of the production machine and monitoring it over time. This can give you information that can be used to improve or optimize your production equipment. These cost savings are difficult to calculate, but is often very interesting to look at.
The data gathered through the machine vision system can in certain situations give information’s that can tell you, that it is time to make preventive maintenance on a machine. Perhaps some parameters are drifting and at a time it could cause defects or a breakdown.
A very important and growing aspect of production, is deep knowledge about processes and production equipment. The discussion about Industry 4.0 is very often mentioned as a new era of production. A machine vision system is and can be an even more interesting part of your Industry 4.0 strategy. Machine vision is gathering many data from the products and through that information about the earlier processes the products has passed. The possibility to have a data feedback from the system, makes it possible to evaluate on the data and to predict or monitor changes. Not only at the moment, but also over time, which makes it possible to see and find new patterns, giving new views on your production and ideas to optimizations or improvements.
Maybe you will be able to classify the defects in more groups than before, enabling you to seek more precise for solutions to specific defects.
A machine vision system can gather Giga Bytes of information – Big data – that can be analyzed through AI (artificial intelligence), to give knowledge that is not possible for a human to see.
Machine vision is an important part of your knowledge in the future.
How do you think about the supplier, who always deliver at time and the wanted quality?
Is he a valuable partner for your business?
Would you discuss price with him as you do with the supplier who always is late and delivers faulty goods ?
No, of cause not.
That’s why a state-of-the-art machine vision system, can help you to get in that position. Your image will be closely connected to the products you are delivering. When you have full control of your processes and the quality you supply, your customers will look at you in a different way – as a partner. Your better image is equal to better earnings.
Finally, some examples where machine vision has been used to gain better and higher earnings.
In glass tube production the JLI machine vision has been installed and made it possible for a known international player to take over the world market, as he was the only one, that could deliver the glass tubes at a very high quality level. He gained customers.
Several known international players in production of medical devices have installed JLI machine vision, enabling them to deliver accordingly to FDA standards – a must for the US market.
In the wood industry a JLI machine vision system has been installed, to secure a high and stable quality inspection, and with a speed that made I possible to decrease the staff. Saving cost and improving quality.
In control of aesthetic presentation of products to the end-user, a JLI machine vision system has been installed. The aesthetic presentation of the products, makes it possible for the producer to get the end-user attention, and thereby get the products sold instead of the competitors’ products. The image of the product was increased through the presentation. Better sales through better image.
You can see some of the challenges, results and benefits from the examples above, in these case stories:
