This application focuses on flexible plastic packaging inspection for defects that can compromise the product’s safety, longevity, quality, and brand.
Flexible plastic food packaging can encounter the following defects:
These packaging defects have severe consequences like consumer health hazards, costly recalls, legal liability, and damaged brand reputation. It is crucial for manufacturers to implement rigorous quality control measures and inspection processes to detect, rectify, and prevent these defects.
But packaging defects can be difficult to detect– they can be variable in type and location. The packaged products themselves can be challenging as well– products can be variable, can overlap or come down the conveyor line inconsistently, and can switch over frequently as manufacturers produce new products.
Traditional machine vision products tend to overkill products when faced with part and defect variability and are slow to reconfigure when parts change over. Manufacturers still rely on manual inspection where machine vision fails. But manual inspection is slow, does not scale to the production speeds manufacturers require, and is subjective to each inspector. Little data is collected with manual inspection, so manufacturers miss production insights and can’t look back at parts and inspection decisions in the event of an escape or recall.
UnitX’s AI-powered inspection effectively detects processed flexible plastic packaging defects where other solutions fail.
First, the OptiX imaging system illuminates and images packaged food products. Then, the CorteX Central AI platform is trained on packaging defects. Lastly, those AI models are deployed to the CorteX Edge inference system to detect and classify defects in-line.
Alternatively, manufacturers can use just CorteX AI if they have existing imaging systems. For example, if a manufacturer wants to detect internal packaging defects, they can deploy just CorteX AI and integrate it with existing X-Ray and CT Scanners for fast deployment of improved defect detection.
OptiX provides superior images that minimize reflectivity while maximizing defect visibility. It has 32 independently controllable lighting sources that can be optimized for various transparent packaging surfaces and defects via software. And its lighting dome design supports a very acute incidence angle of projected light, causing even very tiny defects to cast shadows which increase their visibility.
CorteX accurately detects random, complex defects. It automatically normalizes for variability in positions and orientations and recognizes defects down to the pixel-level. It reduces false positives that lead to scrap and wasted product.
CorteX supports fast AI model development, deployment, and iteration. CorteX AI models are sample efficient– they only require a few images to train on new defect types.
UnitX optimizes yield. In CorteX, can tune quality criteria and visualize the impact on yield before rolling those changes to production. All inspection data is referenceable in one central platform for manufacturers to analyze and identify areas for process improvements and pull up historical records to limit the scope of recalls and fight fraudulent claims when needed.
UnitX provides rapid, 100% inline inspection. OptiX has bright LEDs and fast fly capture speeds of 1m/s for high speed imaging. And CorteX Edge supports high inference speeds (up to 100 MP) to quickly output an OK/NG decision, seamlessly communicating that decision via integration to all major PLC, MES, and FTP systems.
With UnitX, manufacturers prevent food packaging defects and avoid food safety and quality, reduced customer trust, and costly recalls as a result. They automate inspection at the speed of their production to increase food packaging throughput and yield.
In this example, we inspected hot dog packaging for packaging punctures and incorrect product amounts.
Imaging
First, we used OptiX to capture images of the hotdog packages, making sure we captured both defective and OK products.
Training
Next, we used CorteX Central to train our models. We created labels for the two defects: packaging punctures (“puncture”) and incorrect product amounts (“missing_hotdog”).
We then labeled those defects in the images we captured from OptiX, using a low number of images for each defect:
Because of CorteX’s user-friendly interface and the low number of images it requires to train its AI models, it only took us 3 minutes and 42 seconds to complete the labeling for the 2 defects.
Detection
We then deployed those AI models to CorteX Edge to detect defects on new hotdog parts, resulting in thof our two defects.
In this example, we inspected ground beef packaging for food contamination in the seal and packaging scuffs.
Imaging
First, we used OptiX to capture images of the ground beef packages, making sure we captured both defective and OK products.
Training
Next, we used CorteX Central to train our models. We created labels for the two defects: food contamination (“contamination”) and packaging scuff (“scuff”).
We then labeled those defects in the images we captured from OptiX, using a low number of images for both defect types.
Detection
We then deployed those AI models to CorteX Edge to detect defects on new ground beef packages, resulting in the accurate detection and classification of our two defects.
