Machine Vision: Advanced Technology for Industrial Tracking and Tracing
Machine vision finds use in myriad and varied manufacturing applications, but has moved into the spotlight recently for its role as one of the most useful means for managing the tracking and tracing burdens most manufacturers face today. What is this technology, and how does it fit into the scheme of regulatory and quality issues companies must manage?
Machine vision is an important tool for industrial automation used every day in manufacturing plants of all types. It finds use in myriad and varied applications, but has moved into the spotlight recently for its role as one of the most useful means for managing the tracking and tracing burdens most manufacturers face today. What is this technology, and how does it fit into the scheme of regulatory and quality issues companies must manage? Let’s take a look by examining the technology of machine vision, its history, development, and application.
Although admittedly a well-developed and mature technology, machine vision has nonetheless improved immeasurably in recent years. Leveraging the advancements in semiconductor and microprocessor power and in imaging systems, machine vision has dramatically dropped in price and size, while increasing in functionality. Today’s equipment is smarter, smaller, faster, more economical, and easier to use.
Vision systems that were previously large, cumbersome, and expensive now incorporate smart cameras and intelligent sensors and are controlled by sophisticated software running on PCs and microprocessors. Such strides have made the use of machine vision systems on the factory floor simpler to install and suitable for more varied applications. And looking ahead, machine vision vendors anticipate greater miniaturization, extended ease of use, and even lower costs. Beyond these features, machine vision is expected to undergo a higher degree of integration with other automation on the factory floor, in particular with RFID systems. Its capabilities will allow manufacturers to operate more efficiently, minimize time to market, increase consistency, reduce and manage rejects, and maintain component and ingredient traceability that once was lost when raw materials entered the plant.
Four major functions
What exactly does machine vision do? In its simplest terms, it extracts useful information from digital imaging equipment—typically consisting of a microprocessor and a digital camera—for a variety of purposes, including tracking and tracing product, parts, and materials. In an industrial automation setting, such systems can perform a number of functions. Four categories of machine vision applications are primary:
Gauging or measurement
Machine and/or robot guidance
Inspection applications may range from the simple to the complex. A machine vision inspection system in its simplest form may be used to determine whether or not an object is present. Or it may be more complex, such as observing if caps are properly placed on bottles as they move down a production line or if assemblies are complete or labels appropriately positioned.
In gauging applications, the vision system acts as a non-contact measurement device. In a punch press application, for example, holes made in a piece of sheet metal must be precise and consistent. The machine vision system is able to detect those parameters. Are they within acceptable, predetermined, and preprogrammed limits? System measurements reveal if the process is in control. Monitoring hole size is a process control function. Should the tolerance range change, action can be taken well before bad parts are made. Preventive maintenance can be done quickly to bring the production line back into control.
Machine and robot guidance applications harness the power of a machine vision system to determine part location or to control the position or orientation of a machine. A packaging line is a classic example. Items move down a conveyor belt past a vision system which identifies the appropriate part, then directs a machine or robot to pick it from the belt and place it into assembly. The vision system has the ability to determine where the part is and trigger the robotic action of selecting and moving it.
“The machine vision system is essentially closing the control loop,” explains John Agapakis, Siemens’ business manager for machine vision. “It‘sees’ for the robot, closing the visual servo-loop so that the robot can function accurately and properly. The robot goes to where it has been programmed to go; the vision system provides its eyes. The actions may be something very simple or extremely complex. In some cases, the vision system watches a bin of parts to determine the correct part and send the robot to select, orient, and place it. These applications are commonplace today thanks to machine vision technology.”
The final vision system category, identification applications, poses some interesting challenges. Applications may be as simple as recognizing a large object from a small one or separating one color from another. But these applications also encompass such functions as reading a product code using optical character recognition (OCR) capabilities. More complex than reading conventional, one-dimensional bar codes, which are laser scanned, OCR-capable machine vision systems can be used to read a Datamatrix code, a two-dimensional technology developed by Siemens that resembles a small checkerboard. In fact, the only way to read these markings is using an imager and machine vision technology.
Datamatrix codes, standard in many industries, overcome some of the limitations of the traditional bar code. They can be read under low-contrast conditions and read accurately even if partially damaged or destroyed. In addition, their information can be printed on all kinds of surfaces—even one as tiny as the head of a pin. Large amounts of data—up to 2,000 bytes—can be stored in a single code. A Datamatrix code can be applied directly on a part and is unaffected by most environmental elements such as heat or cold. Application of Datamatrix codes on parts, such as Intel’s Pentium chips, helps prevent counterfeiting. These unique markings allow parts to be traced through production and throughout the supply chain, making it easier to isolate a problem should a quality control issue arise or a recall be required.
Expanding the machine vision envelope
An overview of machine vision systems in industrial applications would not be complete without examining their relationship to and interaction with automatic and universal identification technologies such as RFID (radio frequency identification) and UID (universal identification). RFID and machine vision work as complementary technologies for tracking and tracing functions in many industries. Pharmaceuticals offer a good example. Pharmaceutical manufacturers today are seeking ways to track and trace their products to prevent counterfeit drugs from reaching the market and to keep drugs from being used in ways other than for which they were intended (drug diversion). Beyond a desire to self-police their industry, pharmaceutical companies that do business in California are now required under California law (ePedigree legislation) to serialize all products delivered in that state. Other states are expected to follow suit with similar directives.
Universal identification (UID), another track-and-trace initiative, is a mandate of the U.S. Department of Defense that requires any component procured by the DoD having a value $5,000 or more or used in mission-critical applications be identified with a permanent serial number, typically in the form of a Datamatrix code affixed on the part. In these situations, the two technologies—machine vision UID codes on parts and RFID tracking tags on pallets—combine to provide comprehensive systems for tracking and tracing functions such as inventory control, and part and product identification.
Machine vision systems have been part of industrial automation for more than 30 years, and Siemens has been part of the technology since its development as an industry in the early 1980s. Machine vision system applications prior to the advent of microprocessors were limited and consisted of cumbersome, dedicated systems that incorporated a lot of expensive hardware. By contrast, today’s machine vision system is small, compact, and easily and economically applicable, with a smart camera incorporating optics and a microprocessor at its heart.
For more on the inner workings and applications of smart cameras, click here ; for more on how RFID and machine vision technologies work together to facilitate and ensure accurate tracking and tracing, click here .
Case Study Database
Get more exposure for your case study by uploading it to the Plant Engineering case study database, where end-users can identify relevant solutions and explore what the experts are doing to effectively implement a variety of technology and productivity related projects.
These case studies provide examples of how knowledgeable solution providers have used technology, processes and people to create effective and successful implementations in real-world situations. Case studies can be completed by filling out a simple online form where you can outline the project title, abstract, and full story in 1500 words or less; upload photos, videos and a logo.
Click here to visit the Case Study Database and upload your case study.
2012 Salary Survey
In a year when manufacturing continued to lead the economic rebound, it makes sense that plant manager bonuses rebounded. Plant Engineering’s annual Salary Survey shows both wages and bonuses rose in 2012 after a retreat the year before.
Average salary across all job titles for plant floor management rose 3.5% to $95,446, and bonus compensation jumped to $15,162, a 4.2% increase from the 2010 level and double the 2011 total, which showed a sharp drop in bonus.