- This article is filed under:
- » View All
Container Industry Applications of Machine Vision
by Nello Zuech, Contributing Editor - AIA Posted 10/23/2007
Since the inception of the machine vision industry, there has been an emphasis place on the need for applications in the container industry. The earliest addressed the empty bottle inspection application associated with refillable bottles, dating back to 1948. RCA Labs developed the original system using non-imaging electro-optical techniques. Analog video-based techniques were internally developed by Ball Brothers and Owens Illinois to inspect the sidewalls, and Powers Machinery (now part of Emhart Glass) offered electro-optical detectors for the lips of bottles. Today, all of these inspections and more are performed with application-specific machine vision (ASMV) systems.
Some of the earliest applications of digitally-based machine vision were in metal container applications – looking at can lids to make sure they were fully formed, to ensure that there weren’t any crimps, that sealant was properly applied, etc. Today, systems inspect the entire inside of the can and color-based systems can even inspect the decoration on the outside of the container.
Today, machine vision systems can also be found inspecting plastic bottles and preforms for physical properties like shape and visually manifested defects.
Given the widespread use of machine vision in container fabrication, input for this article was canvassed from all known suppliers of ASMV systems addressing applications in the container manufacturing industry. The following were kind enough to respond to our questionnaire:
- Bud Patel, Vice President - Business Development, Applied Vision
- Jorgen Laessoe, President, JLI Vision
- Tom O’Brien, Vice President - Sales, Marketing and New Business Development, Pressco Technology
1. What specific generic container industry applications do your 2D or 3D-based machine vision products address? For which specific materials – plastic, glass, metal? At the container producer? At the container filler?
[Bud Patel – Applied Vision] The majority of our vision systems are used in the container manufacturing industries that include metal, glass, plastic and paper and they are 2D based with throughput speeds in excess of 3000 parts per minute.
[Jorgen Laessoe – JLI Vision] Our main products address hot end glass container production as well as specialised systems for the cold end where stress and special features are detected.
[Tom O’Brien - Pressco Technology] We have been providing primarily 2D inspection system solutions since the early 1980’s with a focus on bringing state-of-the-art technology to the metal, plastic, and glass container and closure markets. Our deep experience in the industry facilitates robust inspection system solutions for domestic and international container manufacturers, closure manufacturers, and container fillers.
2. What are the benefits the container industry derives from the use of your machine vision-based products?
[Jorgen] The main benefit is fast feedback to IS machine control thereby correcting faults that would otherwise only be found much later in the cold end. The fast response time can give an improved pack rate of 2 - 3%. This gives a payback time on a typical line of 5 months. Also the systems detect critical faults, like bird swings, and reduces the likelihood of these faults reaching the customer with a factor of 500 or so.
[Tom] In its most basic role, the vision system is used as a go/no-go gauge to make a pass/fail decision based on the presence of defects, insufficient fill level, etc. In addition, our systems are designed to take inspection to another level by identifying a machine component that produces a repetitive defect. For example, in the metal container industry, our systems can identify a Body-Maker ID on a beverage can to allow traceability to a specific Body-Maker in the manufacturing process. In a stretch blow molder, our patented correlation feature can identify a repetitive defect coming from a mold, spindle, or transfer arm. With this process-monitoring capability, our systems quickly identify the source of a problem in a manufacturing line to speed trouble-shooting and minimize spoilage.
[Bud] There are two important fundamental benefits; 1) eliminating functional, safety, and appearance defects in their manufactured products and 2) receiving timely process feedback for continuous process-monitoring and improvements.
3. Can you briefly describe your machine vision-based solutions for the container industry? Specific type of machine vision hardware (frame grabbers, smart cameras, embedded vision processors, etc.), lighting arrangement, camera details? And, why you selected the specific components in your designs?
[Tom] We supply a variety of machine vision-based solutions for the container industry. These include a variety of both analog and digital based systems, as well as smart cameras, depending on the application requirements. Also, depending upon the application, we use combinations of both front lighting and back lighting in various configurations based on our CHROMAPULSE™ patented multi-color, multi-directional illumination. In summary, we supply a complete solution that is designed to solve a specific problem or set of problems for our customers.
[Bud] Our systems are comprised of off-the-shelf high performance industrial-grade computers, digital Gigabit Ethernet (GigE Vision™) cameras, some proprietary part-tracking and illumination hardware packaged for challenging industrial environments.
[Jorgen] The computers are industrial XP machines with frame grabbers and I/O modules. All systems are connected to the Internet and can be operated from our office in Denmark. Lighting is custom built for the specific application. At present, we use our Dynamic Light Box for backlighting applications, which is a powerful LCD flat screen where the vision computer generates patterns, stripes, dots, black and white contours, and even colour patterns, to enhance specific defects in transparent products.
4. What new products or advances to existing products have you introduced in the past year or so for applications in the container industry?
[Bud] We developed groundbreaking technology comprised of new illumination systems, and software to provide substantial process information upstream to help correct the can manufacturing process and reduce spoilage rates. We call this technology VTRAC™. This technology allows the manufacturer to reduce their spoilage rate while increasing their throughput.
[Jorgen] A new system will be launched later this year. We cannot say much about it at this moment.
[Tom] Pressco Technology has introduced RETRO-SPEC™, our patent pending easy-to-use and intuitive user interface. We’ve also launched our next generation digital-based system and new process monitoring capabilities with Body-Maker ID for metal beverage cans, spray dot identification for metal beverage cans, and mold number ID for plastic beverage closures. In addition, we have pioneered new capabilities in entering the filled container market segment for cans, glass, and PET.
5. What advantages have these advances produced for the container industry?
[Tom] Our intuitive new RETRO-SPEC™ user interface is a breakthrough technology, which minimizes system set up time. For the first time, operators can quickly optimize the inspection parameters based on historical inspection data. The new process monitoring capabilities mentioned above are a continuation of a long line of developments, dating back to the early 1990’s, which make our vision systems an integral part of a container manufacturing or container filling line. As a supplier also to the filled container market, Pressco Technology offers the broadest inspection capability on PET and can filling lines in the industry today. This allows a container filler to standardize on one supplier of solutions for all of their inspection applications on a filling line.
[Bud] We were the first company to use digital (GigE Vision™) cameras for our customers in the container industry (2003). The GigE Vision™ camera technology provides high-speed image-data communication using common and inexpensive cables and connectors. The benefits to our customer base include high performance imaging over long cable runs. The use of digital cameras provides a more noise-immune imaging chain, thus providing a better image quality, and therefore, better inspection.
Regarding VTRAC™, we were the first to successfully implement this for the two-piece beverage can applications where we read the body maker identification code on the bottom of the can (low-contrast, randomly oriented numerical code) and correlated body and neck defects to the appropriate body maker and necker-pocket. With this critical information at hand, the can manufacturer can quickly solve the problem upstream and reduce their spoilage rate while increasing their throughput.
6. What advances associated with the technology infrastructure of your products made those new products or advances possible (optics, lighting, vision hardware, vision software, cameras, etc.)?
[Bud] GigE Vision™ standardization and digital technology enabled our evolution to GigE Vision™ digital cameras. Faster microprocessors enable more sophisticated algorithms and more cameras supported by a given processor. Various plant networks and SCADA interfaces allow for key process feedback from the vision system upstream, as mentioned earlier.
[Tom] With a greater emphasis on process monitoring there is now a greater need, not only to identify the source of the defect, but also to communicate real time data to customers’ plant monitoring computers. In the case of reading a Body-Maker number at line speeds approaching 3,000 cans per minute, we developed a very comprehensive solution, including new optics, lighting and imaging algorithms. In order to communicate the production statistics, we have implemented OPC to allow this information to be transferred to our customer’s plant monitoring computer systems. To address the range of filling applications, no new core technology was required, however, we needed to make substantial changes to product packaging and develop new software algorithms.
7. Have the machine vision applications your company addresses in the container industry become easier or harder over the years (for you as the solution provider and for the customer)? Why? How has this impacted the price a customer pays?
[Jorgen] JLI makes the complicated high-risk projects where others dare not go (or are wise enough to avoid). We never do the simple jobs. Therefore, we have a constant challenge to stretch the technology and refine the methods. Turnkey systems involves all aspects of the solution, including mechanics, lighting, optics and the vision computer loaded with all the necessary software. The lighting is an especially important component and we have developed very powerful lighting systems for our glass tube inspection systems. These illumination units are not commercially available.
[Tom] The applications in the container industry, from a requirements standpoint, have either remained the same or become more difficult over time. For example, the basic inspection done on the inside of a two-piece beverage can is basically the same today as it was ten years ago. However, customers now want to identify the Body-Maker number on the bottom of the can to identify the source of a potential defect. This is a new application that is more demanding and that has evolved over the past few years. Generally, on those applications that have remained constant in their requirements, customers expect to pay less based on technology evolution in vision systems.
[Bud] Like most industries, the container industry is always facing new challenges in a highly competitive environment. Customers are producing containers at higher speeds (exceeding 3000 containers per minute) and are looking to find smaller defects than ever before. They are very cost-conscious and cannot afford any false rejects, which puts added pressure on the vision vendor. However, as mentioned before, the technology improvements are assisting us in providing better solutions for our customers. One aspect that continually improves is the user-interface. Through better software and display technology the systems are easier to set up and maintain. The overall system costs are declining but at slower rates as compared to the machine vision industry in general.
8. What changes in the underlying 2D or 3D machine vision technology (vision engines, lighting, cameras) do you anticipate in the next 2 – 3 years?
[Tom] We expect to be delivering solutions that allow more cameras and application-specific lighting modules to be installed at various locations in the manufacturing process where value-added operations occur. The technology will continue to evolve with the development of faster processors, higher resolution digital and color cameras, and smarter algorithms to lower the manufacturers’ costs, reduce their spoilage, and increase their line speeds.
[Bud] In the past, vision systems were regarded as “stand-alone at the end-of-the-line” devices. In the near future the vision systems will be more “integral” to the process by monitoring and inspecting to help maximize the overall efficiency and quality of the manufactured products in the container industry.
[Jorgen] Everything gets faster and smarter. For JLI, the price of our systems increase as we take on more and more demanding applications.
9. How will those changes impact container industry applications?
[Bud] These new generations of intelligent vision systems will allow can manufacturers to absolutely minimize spoilage, reduce downtime, and improve maintenance procedures.
[Tom] These enhancements will allow some of the more difficult applications to be done, which until now, have only been marginally acceptable to customers. An example would be full inspection of the outside of a decorated metal beverage can. In addition, more capable systems will result in fewer false accepts and rejects, which will open up new applications and lead to a greater acceptance of machine vision technology.
10. Are there still barriers to the adoption of machine vision technology for container industry applications and what are they?
[Jorgen] It is a problem to make operators in the hot end use advanced equipment. They regard glassmaking as an art and not as a production process that is open to automation like any other industrial production. With younger generations moving into the hot end, the acceptance of high tech equipment for glass-making will improve.
[Tom] Customers in the container industry want identifiable paybacks for their investments made in inspection technology.
[Bud] The barriers to entry into the container industry remain specific application knowledge of the can-making process and applying this knowledge to lighting and optics, software algorithms, and user interfaces.
11. Are there market/process changes that are taking place in the container industry that are driving the adoption of machine vision?
[Tom] Probably the most significant changes that have occurred are associated with the light-weighting of containers to reduce material cost. We have seen this occur in the metal container industry and are now experiencing light-weighting in the plastic container industry. In addition, there is greater interest and acceptance of machine vision systems for process monitoring and for closing the loop on manufacturing machines or lines.
[Bud] As mentioned earlier, the vision systems today are viewed as an integral part of the can-making process by monitoring and inspecting to help maximize the overall efficiency and quality of the manufactured products in the container industry.
12. What advice would you give a prospective customer for a machine vision-based solution addressing container industry applications?
[Jorgen] Work out the payback time and make the operator part of the automation task. Give a bonus to the staff when they use the systems and improve the productivity.
[Tom] To work with an established supplier and one that has relevant application knowledge in the container industry.
[Bud] It is critical that the vision vendor understand the container manufacturing process and has developed an ASMV solution for these specific applications. Another important factor is the amount of experience the vision vendor has in the container industry.
There are currently no comments for this article.
Leave a Comment:
All fields are required, but only your name and comment will be visible (email addresses are kept confidential). Comments are moderated and will not appear immediately. Please no link dropping, no keywords or domains as names; do not spam, and please do not advertise.