In the meat industry, new methods are being invented regularly to deal with the conflict between increasing production volume and quality control (QC). Vision systems in particular are on the cutting edge of quality control measures.
A vision system can do any number of things, but generally, they use a camera to take images of a product and make a QC decision based on a physical trait. Because they are based entirely on images, they can run into problems with lighting and color.
To demonstrate these problems, we could consider the example of grading a beef carcass based on the quality of the ribeye cut. The ribeye is used to grade the whole carcass by industry and USDA standards, so this is an important process.
There are a number of different properties to look for when grading the ribeye. Fat content/marbling is an important factor in choosing whether the meat can be considered prime, choice, or select. For certain cattle, like the renowned Kobe, this is especially important for getting top dollar.
Fat detection can be harder than one would think, as it takes precise lighting, camera angle, and camera quality. The system needs to be able to pick it up and try to match it with a database of accepted images to see which image most closely resembles it and what grade that image ultimately was assigned. This image computation happens very quickly and can be automatic.
If the lighting in the area is dim or in a certain hue, the color contrast between meat and fat can be dulled, making it difficult for a system to pick up. To solve this, some places use ultra-violet (UV) light because it makes the color white pop out in great contrast. However, other colors may be harder to determine when UV is used. The right type of lighting depends largely on the tests needing to be made.
If a test for color is required, you must use some sort of light that is very color neutral and provides high contrast. With the ribeye, this can estimate the redness of the meat to assure that the carcass is healthy and fresh. The only trouble is that color can be deceiving in certain circumstances. If a beef carcass hasn’t been exposed to the air long enough after being opened up, its meat will have a much darker hue which could be misleading to the vision system if the carcasses passed through QC too early.
Installing a vision system takes very careful and precise planning, but when all of the bugs have been worked out, it can be an invaluable asset to any plant.
Vision systems are helpful in the sense that they allow you to do visual checks automatically without employee inspection, but they can be difficult to set up ideally.
As an example, we could see how a vision system might apply to selection of hams. When cutting the pig into its individual components, the ham is typically cut right above a joint where the aitchbone is located. The distance from the cut to the aitchbone relates directly to the diameter of the ham bone at the cut. This can be used to make a general comparison of ham sizes.
The five different muscle groups in a ham can be used for a variety of things. Certain sized hams get put into machines that cut them into spiral cut hams and some are sold as whole hams.
To categorize each ham for the proper use, weight and size are needed. This could be done with the combination of conveyor scale and vision system. While the ham crosses over the conveyor scale, a camera can look at the cross section to detect the size of the bone.
This is where the process can become difficult. Vision systems can run into problems with lighting and color that you might not encounter with human handling. For example, a dirty bulb which discolors the light could in turn discolor the meat for the camera. This makes it very tough for the camera to identify things correctly because vision systems compare the product to a number of acceptable images to justify what range the product falls into.
On the color side of things, the process can become difficult based on where the ham was cut. The vision system detects pixels that vary from the red hue of the meat in order to find the bone, but the color of the bone isn’t always pronounced. If the bone is cut in such a way that it is less visible, the vision system could miss the bone entirely as the marrow itself often contains colors similar to the ham.
While this operation has a lot of potential, it requires that the cutting of hams be precise for bone visibility and that lighting be optimized. On the other hand, a human could spot the bone fairly easily after a little examination. This shows that both human estimation and machine vision have a place. Vision systems have the power to greatly speed up certain processes, but the current technology for them does much better with well-defined trait differences. Feel free to ask questions about some of the processes that machine vision could be beneficial for.
In recent years, packaging equipment has been increasing in throughput to the point that it is not feasible to have someone manually inspect your products with satisfactory results.
Line speeds have just become too fast. There fore you are left with two choices. The first is to inspect a random sampling of your products and hope that it is representative of all the rest. The second choice is to inspect each and every single product that goes across the production line with the use of a machine vision inspection system that does not get fatigued or whose decisions won’t change based on the mood it is in. A machine vision inspection system will inspect every single product that goes across the production line based on the same criteria every time!
Machine vision technology has advanced rapidly in recent years, bringing the price down, while at the same time increasing its capabilities. We have been in the machine vision business for more than 20 years. That allows us to bring some options to the table that other suppliers cannot.
Paired with our expertise in the automated weighing field, we can combine aspects of automated weighing and machine vision to provide you with a checkweigher that can also do vision inspection. Let’s use a frozen pizza manufacturer for an example. Today, they are making sausage pizzas, and in order to be in tolerance, a sausage pizza is to weigh between 450 and 465 grams. Our checkweighers can easily determine if this pizza is in spec as far as weight, and so can many other manufacturers. But, how many of those other manufacturers can also assure you that the sausage pizza is in a sausage pizza box and not a pepperoni pizza box, and that the barcode on the box can be read by a scanner? We can also protect your brand by making sure that all the colors are absolutely correct and that images are not blurred.
Another good application where machine vision and scales work well together is in an environment where several different products are produced at the same time. This may require more than one checkweigher to run the different products or an operator to be stationed at the checkweigher to change the product number as the different products come to the scale. When coupled with a good machine vision system, the system could be trained to automatically change the parameters in the checkweigher.
By putting these two inspection systems together, you really get a complete inspection system that has many benefits. Because the two systems will share some common components like conveyors and controllers, you will wind up with a much smaller footprint saving some valuable floor space, and costs can be brought way down due to eliminating the redundancy in components.
Some of you may be wondering: “What can I automatically sort, grade, or classify with the equipment that is out there today?”
With the type of equipment that we offer, we usually think of sorting by weight. We are capable of weighing anything and most of the time we can do it in motion. Today, even semis are capable of being weighed as they travel down the road at 70 MPH. In many places, items are being weighed and check-weighed by the gram and less. Once the user has a given weight, sorting is possible and can be done quite easily.
Now let’s add grading.
Grading is the term where we are adding another factor to sorting. Yes, you can grade by weight, but what if you need to grade by color or quality? Systems are being developed each and every day with an in-motion conveyor scale and a vision system that has the ability to grade and sort.
Vision systems have the ability to see color, to see the marbling in a piece of meat, to determine the thickness of the given object, and much more. Vision is being used to do Q.A. on just about every type of application there is.
When we sort and/or grade with vision, you need software that has specific parameters on which to measure what classification the product falls into. We then will know what needs to be done with the product being graded. If you enter the weight back into the equation, you have another set of parameters to work with. If you have done your set up properly, weight and grade will fall into specific classifications.
There is one other way of grading that is being done today in a number of applications, “human grading,” with the assistance of technology. Vision camera’s today have a tremendous amount of ability; however, they do not decipher information in the same manner as a human does. There are systems in place today where we have developed touchscreens giving the grader access to just touch an area on the touchscreen indicating a change in grade or an abnormality to the specific product area.
Add in a weight and now with proper software you have as many classifications as you need.
The plant manager of a small-sized hog slaughter facility wanted to implement a Grade Yield Program.
Among his goals were to reward producers for animals that fit the ideal weight range, while avoiding the problem of trimming a bad shoulder or loin on an over-weight animal, only to have that animal wrongly qualify for a premium pay out.
But like so many operations today, this plant manager’s IT department consisted of someone who could keep the copy machines and personal computers running.
Software programmers are simply too expensive for most small to medium sized operations to keep on staff.
That’s where we came in.
Our Programming Services Department gave this plant the ability to document which part or parts of the animal were condemned and trimmed with a keyboard entry near and interfaced to the Hot Scale.
This program prevents undeserving animals from backing into the Premium Payment category.
Additionally, it gives the producers a clear explanation for the payment they received and does so on all of the different contracts that the plant has with its various producers.
The producers have told the plant manager how much they appreciate the clear reports.
Now everybody knows and understands their payments and the plant is more efficient.
While we deal with all sizes and needs of plants, often our talents can be most utilized in some smaller plants.
The industry knowledge coupled with system manufacturing, and system adaptability allow us to customize the solution to each application.
Ask us how our staff of full time software programmers can help your plant run more efficiently!