Label and print inspection for production lines
Label presence, position, SKU and print checks for bottles, jars, cans and cartons on South African packaging lines.
Technical review: Frank GuoReviewed
A camera station can check that every label is present, straight and belongs to the SKU being run, and that printed text is complete. It works when the label is visible from a fixed position and lighting is controlled. It cannot see labels hidden by shrink-wrap folds or confirm adhesive strength. We verify this on your samples first.
What a label inspection station checks
Four checks cover most label faults. Which ones you need, and how tight the limits are, is set during the sample test.
Label presence & position
Missing, crooked, shifted or doubled labels, checked against a reference position taught from good samples.
Wrong-SKU labels
Reads the label barcode, or matches printed text and artwork, against the product selected for the run, so a wrong label after changeover is caught.
Print completeness
Missing or broken print, smears and incomplete text in defined areas of the label.
Wrap-around labels
Seam position, overlap and lifted edges on round containers.
Where label inspection stops
We set these boundaries before a project starts, so nobody expects the camera to do what it physically cannot.
Not detectable with a camera
- Adhesive strength. A camera sees position and appearance, not how well the label is stuck.
- Hidden print. Text or artwork covered by an opaque sleeve or folded shrink film cannot be inspected.
- Freely rotating containers with one camera. A single view cannot see a label on a bottle that arrives at any angle; this needs orientation or several cameras.
Works reliably when
- Lighting is controlled. The station's own lighting, shielded from daylight, keeps images consistent across shifts.
- Products arrive consistently. Stable spacing and presentation, with a trigger sensor, and an encoder where speed varies.
- Defects are defined. An agreed list of faults and reference samples, so limits are set on evidence.
How a label inspection station is laid out
Plan view of a typical single-sided station. The real layout is designed around your line after the site survey.
Swipe sideways to see the whole station →
What a pass and a reject look like
Illustrations of the three most common outcomes. Limits such as how much tilt is acceptable are agreed with you during the sample test.
Label correct
Label present, straight within the agreed limit, print complete, and the barcode matches the product selected for the run.
Label crooked
The label is rotated beyond the limit agreed at the sample test. The product is tracked to the reject point and removed.
Wrong product label
The code read from the label belongs to SKU B, but the line is running SKU A. This is the typical fault after a changeover.
What we need to know about your line
These conditions decide the station design. None of them has a fixed answer; each is confirmed on site or during the sample test.
| Condition | What we need to know | Why it matters | Confirmed during |
|---|---|---|---|
| Line speed | Units per minute or per hour, and whether the speed varies. | Sets exposure time, lighting power and the trigger method. Variable speed usually needs an encoder. | Site survey |
| Product spacing | Gap between products and how stable it is. | Each product must be imaged and rejected on its own; touching products need spacing or a different reject. | Site survey |
| PLC and controls | PLC make and model, spare I/O or network ports. | Decides how pass/fail, reject timing and SKU selection are exchanged with the line. | Design review |
| Reject mechanism | Existing reject or not; compressed air available. | Failed products must leave the line reliably; a confirmation sensor proves they did. | Site survey |
| Changeover | Number of SKUs and how the running product is selected. | Each SKU needs a recipe; automatic selection from the PLC avoids operator error. | Sample test |
How the station is accepted
Two numbers matter, and they pull against each other. We define both, and how they are measured, before FAT and SAT.
False reject rate
The share of good products the station rejects. Measured on an agreed set of good samples at line speed; the acceptable level is agreed per project.
Missed defect rate
The share of defective products that pass. Measured with defect samples fed at known positions; for critical faults such as a wrong SKU, the target is set with you.
Reference samples & recipe control
Reference samples and recipe settings are documented and access-controlled, so limits do not drift after handover.
What to prepare for the sample test
Before we propose a station, we check your labels on real samples. This is what we ask for; we confirm quantities and shipping with you first.
Sample preparation guideGood samples
Products that pass today, covering the normal variation: print runs, colours, gloss and matte finishes.
Defective samples
Each fault you need to catch, including borderline cases: crooked, missing, wrong or badly printed labels.
Product information
Container and label drawings or photos, and the list of SKUs that run on the line.
Line information
Line speed, product spacing, PLC make and model, and whether a reject mechanism is fitted.
Label inspection questions
How does the station handle glossy or metallic foil labels?
What happens at a product changeover?
Can the same camera read the date code as well as the label?
Does a label inspection prove the label is stuck on properly?
Request an assessment for this application
Describe your labels and your line, and send a few good and bad samples. We will tell you what a station can check and what it would involve.