Researchers field test new plastic detector
in their products, if trials of a new machine that detects coloured
objects prove successful.
The new system, now in final development stages, is scheduled to begin field testing later this summer at the Georgia Tech Research Institute(GTRI). the machine would help food processors in their battle to maintain the safety of their products. Consumer demand and an increased focus on food safety through regulator oversight andregulations is a driving trend in the market.
Many food processors use detectors to keep metal fragments from ending up in finished products. However these detectors can't identify plastic and other foreign objects.
GTRI's new machine, being developed by research engineer John Stewart, is a computer vision system that will automatically detect and then remove colored foreign objects from the food stream.
Stewart began building the system in response to the increasing use of plastic at food plants. As plastic becomes more widespread, used in everything from conveyor belts to latex gloves,plastic contamination becomes a growing concern for many food processing operations.
"Incidences of plastic contamination are infrequent, but when they occur, the fallout can be extensive," Georgia Tech stated in a press release. "Recalls are expensive, notonly in terms of logistics and returned product, but also because recalls can tarnish a company's brand image and reduce consumer confidence."
Stewart's came up with an inspection tool using computer vision technology coupled with sophisticated colour discrimination algorithms. The system sits above a production line adjacent to metaldetectors.
Controllers first train the machine to identify the conveyor belt being used and the desired characteristics of the food product. As the product moves along the conveyor, the computer-vision systemcaptures digital pictures and analyses them.
If the system sees an object it doesn't recognise, it records the digital image and activates an alarm and a device that automatically removes the product from the line.
The system can determine a full range of colour. However Stewart's team have so far focused on finding blue and green objects. Blue has become a standardised colour for plastic used in foodprocessing plants.
"Few foods are blue, so food processors hope that line workers will recognise any foreign objects making their way into the product stream," Stewart stated in an announcement abouttesting the machine.
GTRI's new machine would replace human inspectors, who may miss the plastic contamination in a product since the processing line is moving so fast. On average products are moving at about 12 feetper second, or eight miles per hour, on a processing line
"If a person blinks or looks away for even a second, they can miss a problem," he said. "In contrast, machine vision is very diligent. It doesn't get tired or bored. What'smore, line workers see only the top of finished products. GTRI's computer-vision system captures additional views of surface area by taking digital images as products tumble off one conveyor belt andonto another.
In lab tests, the system was able to identify foreign objects as small as 1.5mm with few false alarms and close to perfect accuracy, the researchers said. The field tests will determine if suchresults are achievable under processing plant conditions.
The system is designed to operate on conveyor belts moving 12 feet per second. In the lab, top conveyor speeds were set at three feet per second. The researchers also simulated factoryconditions by using dimmer lights.
The ultimate goal is to make the computer-vision system as fast and accurate as possible without outpricing the technology for industry users, Stewart stated. GTRI has teamed up with Gainco, anequipment manufacturer in the US.
Although the lab tests focused on finding plastic fragments in poultry products, GTRI's computer-vision system can also identify non-plastic contaminants, such as glass. It can be used for meat andother food products.
"We're trying to make the system as generic as possible, so anything that doesn't look like the product will be detected," Stewart stated.
