Yoghurt cup lid presence and alignment
A yoghurt cup can have a lid present and still have an unacceptable visible seal margin. Compare the lid edge with the cup reference.
Line solution
Test this on your samplesYoghurt cup inspection: lids and date codes
// Short answer
Yoghurt cup inspection is a camera check of every cup after the sealing station: it confirms a lid is on and seated, reads the best-before date and batch code on the lid, and compares them with the run. Failed cups are rejected by the filler, lane by lane. addaScan builds it on a HIKROBOT vision sensor over the lanes, or on two HIKROBOT cameras with a vision controller. An indicative installed budget for a typical application is R 110 000 – R 680 000 excluding VAT.
// At a glance
The station in three lines
- Faults caught
- Foil lid missing or lifted; date or batch code missing, smudged or wrong; per lane
- Configuration
- HIKROBOT vision sensor over the lanes, or top and side cameras on a VC3000 with dome light
- Line link
- Per-lane reject signals to the filler; expected date and batch from the filler or PLC
// Fit & limits
What this station will not do, and what it needs
What it will not do
- Seal and leaks: a lid that is on and flat may still be badly sealed; seal strength and leaks are not visual checks.
- Contents: fill level, mass and what is in the cup are not checked.
- Code grading: the date is read and compared, not graded to a print-quality standard.
- Under the lid: nothing below the foil is seen.
What makes it work
- Samples: good cups, cups with no lid, a lifted lid and bad date codes, on your foil.
- A filler signal: an index signal per row and a per-lane reject the vision result can drive.
- Expected values: date and batch for each run from the coder or the PLC.
- A washdown plan: housings or lens caps fitted, and cleaning that does not aim at the camera.
// On South African lines
Yoghurt cup inspection on South African dairy lines
South Africa has many dairy processors, not only a few large ones: in December 2024, 129 of the persons and institutions registered with Milk SA were processors, and the fermented products declared to Milk SA include maas, yoghurt, kefir and buttermilk (Milk SA, 2024). Yoghurt and maas are often filled into cups and sealed with a foil lid on a cup sealing machine; fresh milk goes into bottles.
The dairy regulations set what the pack must show. Each container must carry the best before, use by or sell by date and the batch code or number, and the markings must be clearly legible, with lower-case vowels at least 1 mm high (R260, 2015). The food hygiene regulations require prepacked food to be packed so that it cannot be removed without the lid or seal being removed or damaged (R638, 2018). The rules set the outcome; a camera checking every cup is one way to show the lid is on and the code is there, cup by cup and lane by lane.
A lid that is on and flat is not proof of a good seal. This station sees the lid and the print; seal strength and leaks need the sealer’s own controls and your QA tests.
What the line reports, and what yoghurt cup inspection checks
| What the line reports | What a station checks | Where it stops |
|---|---|---|
| “A cup came off the filler with no foil lid” | Foil lid present on every cup in every lane, from above. | A lid that is there but not sealed looks the same as a sealed one. |
| “The lid is on but lifted on one side” | Lid lying flat on the rim: from above on the single-sensor station, and from a side view on the two-camera station. | A small lift that does not show in the image is not found; leaks and weak seals are not tested. |
| “The best before is missing or smudged on the foil” | Date and batch read on each lid and each character checked against the expected string. | Print on embossed or creased foil needs the right light; it is proven on your lids in the sample test. |
| “After the flavour change the date was still from the last run” | The read date and batch compared with the values sent by the filler or the PLC for this run. | The camera checks against the expected values it is given; a wrong value at the source is not found. |
| “We also need the date checked on the milk bottle line” | The same OCR check on the cap or bottle shoulder, as a separate station with its own recipe. | A milk bottle line is a different station and a different budget from the cup filler. |
// The complete solution
Cup lid and date code station: hardware, software and installation
A camera above the filler’s outfeed sees all lanes at once after the cup sealing machine. The vision system checks each cup position for a flat lid and a readable, correct date code, and sends a per-lane reject signal to the filler.
[Schematic]Plan view, not to scale
Swipe sideways to see the whole drawing →
[1]: The filler’s index signal tells the camera a row of cups is in place.
[2]: Top camera or vision sensor with a dome light checks each lid and reads each date code.
[3]: Side camera (two-camera station) sees whether a lid is lifted at the rim.
[4]: The filler’s own reject removes failed cups, lane by lane.
[5]: Confirms the rejected cup left its lane.
Hardware
| Component | HIKROBOT example | Role in this station |
|---|---|---|
| Vision sensor (single sensor) | HIKROBOT SC3000-series 5 MP vision sensor with OCR over the lanes (budget priced on MV-SC3050M) | Checks every lid and reads every date code across the lanes |
| Cameras (two cameras) | 2 × HIKROBOT MV-CS200-10GM 20 MP mono GigE cameras with 25 mm lenses (MVL-HF2524M-10MP), each in an IP67 camera housing | Top view of all lanes for lid and code; side view for lid seating |
| Processing (two cameras) | HIKROBOT VC3000 vision controller (e.g. MV-VC3302P) running HIKROBOT VisionMaster | Runs the per-lane checks and sends one result per cup |
| Lighting | HIKROBOT MV-LRDS-H ring light (single sensor); a dome light, such as HIKROBOT’s MV-LBES-H flat dome, and an MV-LLDS-H bar light (two cameras) | Even light on shiny foil so the print is readable and glare does not hide a missing lid |
| Trigger | Not HIKROBOT: the filler’s index or cycle signal | Fires the camera once per row of cups |
| Reject | Not HIKROBOT: the filler’s own per-lane reject, driven by the vision result | Removes failed cups without a separate reject |
| Washdown protection and panel | Local supply: stainless mounts and shroud, IP67 cable glands, washdown protection, panel | Keeps cameras and lights working through cleaning |
Software and vision tools
| Tool or software | What it does here |
|---|---|
| OCR and character comparison | Reads the date and batch on each lid and compares them with the run. |
| HIKROBOT VisionMaster OCV | Verifies every printed character against the expected string, so a broken digit fails. |
| Presence and edge tools per lane | Lid present on each cup; lid edge position at the rim in the side view. |
| Deep-learning character-defect check (option) | HIKROBOT shows deep learning finding broken, distorted and smeared inkjet dates on fresh-milk packs; priced after the sample test if your print needs it. |
| Lane recipe | One recipe per cup size and lane layout, selected by the filler or PLC at changeover. |
Installation and integration
01
Mounting and enclosure
Camera above the filler outfeed on stainless mounts that clear the cleaning area; IP67 housings or lens caps fitted, because HIKROBOT’s IP67 ratings depend on the lens cap or housing being in place.
02
Lighting and shrouding
A dome light over the lanes so foil, which mirrors any point light, is lit evenly; a shroud keeps plant lights and window light out.
03
Trigger and tracking
The filler’s index signal triggers one image per row; each cup position in the image is tied to its lane.
04
PLC link
One pass or fail per lane to the filler’s reject over digital I/O or the filler PLC’s fieldbus; the expected date and batch come from the coder or the PLC.
05
Reject and confirmation
The filler’s own lane reject, with a confirmation that the cup left; if the vision system is not ready, the filler holds or rejects, it does not pass cups unchecked.
06
Recipes and operator screen
Cup size and product recipes switched at changeover; a screen shows the last rejected cups with lane and reason.
// Indicative cost in South Africa
What this station costs, excluding VAT
Indicative budget for a typical application of this kind. Every project involves custom requirements (your products, line speed, layout, PLC and acceptance criteria), so the final price differs from case to case. A firm price follows the sample test.
- Low
- R 110 000
- one sensor, few lanes
- High
- R 680 000
- two cameras, wide filler
- Basis
- Typical application of this kind, indicative
- South African rand, excluding VAT
- Prices as at 2 October 2026
- Travel (Gauteng): outside Gauteng quoted separately
// Included and excluded
The low figure is one vision sensor over a narrow filler that checks lid presence and date code presence. The high figure is two cameras on a controller for a wide filler, reading each code and checking lid seating from the side, with per-lane signals to the filler’s reject. No public price exists for stainless washdown enclosures, so the high figure carries an allowance for them. Deep-learning print checks, changes to the filler, utilities and VAT are excluded.
Hardware is priced from public list prices converted at the ECB reference rate, plus freight, clearance and margin; engineering at R 900 – R 1 300 per hour. How these ranges are built →
Swipe the table sideways to see the high figure and the basis →
| Cost block | Low | High | Basis |
|---|---|---|---|
| Vision hardware (landed, with margin) | R 50 100 | R 192 700 | Low: one 5 MP vision sensor over 2 to 4 lanes, ring light. High: two 20 MP cameras with lenses for a 6-lane filler, dome light, long bar light, VC3000 controller, two IP67 housings |
| Software licence | R 0 | R 40 100 | Vision sensor software is free; high case with the HIKROBOT VisionMaster standard licence |
| Local materials and reject | R 15 000 | R 90 000 | Low: washdown-tolerant bracket, IP67 cable glands. High: stainless mounts and shroud, allowance for washdown protection, panel; the filler’s own reject is used in both |
| Engineering (recipes, PLC link) | R 27 000 | R 182 000 | 30 h at R 900 to 140 h at R 1 300 |
| Installation, FAT, SAT, training, documents | R 14 400 | R 93 600 | 16 h to 72 h |
| Project management | R 5 400 | R 39 000 | 6 h to 30 h |
| Travel (Gauteng) | R 1 500 | R 6 000 | Outside Gauteng quoted separately |
| Total (rounded) | R 110 000 | R 680 000 | Low: one sensor, few lanes. High: two cameras, wide filler |
// Related technology
Which technologies this scenario uses, and why
The technologies behind this station, how much each matters here, and where each is explained in depth.
| Technology | Role | Why it matters in this scenario | Read more |
|---|---|---|---|
| OCR/OCV of printed codes | Core | Reading the best-before date and batch on every lid, and checking them against the run, is the core task. | OCR/OCV inspection |
| Area scan cameras with a vision controller | Core | A wide multi-lane filler needs high-resolution cameras on a controller to see every cup at once. | Camera and lens selection |
| Vision sensor (HIKROBOT SC1000–SC3000) | Supporting | On a narrow filler one vision sensor covers the lanes without a PC. | Choosing a smart camera |
| Lighting technique | Supporting | Foil mirrors every light source; a dome light is what makes the print and the lid edge readable. | Machine vision lighting |
| Presence detection and counting | Supporting | Each cup position is a presence check for its lid. | Presence and counting |
| Deep learning (classification, segmentation, DL OCR) | Supporting | Smeared or broken inkjet dates on foil are a deep-learning character-defect task if rule-based OCV is not enough. | Rule-based vs deep learning |
| PLC link and reject with confirmation | Supporting | The result must reach the filler’s lane reject for the right cup. | PLC and reject integration |
// Acceptance
How the station is proven before handover
Every station is accepted on your products, against numbers agreed in writing before the build. How we deliver →
01
Sample test
Your cups with good lids and codes, and with no lid, a lifted lid, a smudged date and a wrong date, imaged under the proposed light on your foil; the result says what is reliably found and what is not.
02
FAT with challenge cups
Before delivery, marked challenge cups in every lane must each be rejected, and good cups in every lane must pass.
03
SAT on the line
On your filler, at production speed, the challenge cups are run again in each lane, and false rejects and escapes are counted over an agreed run of production.
Keep cup-filler results associated with each lane
A cup filler may present several lanes in the same image. The decision needs the lane, cup position and recipe as well as a global image result.
| Decision or event | Check or action | Acceptance boundary |
|---|---|---|
| Lid presence | Inspect each occupied lane separately | An empty lane must follow its defined machine state. |
| Visible edge lift | Test actual foil reflections and normal curl | Appearance alone does not prove bond strength. |
| Code | Check required product/date/batch fields by lane | A readable date can still be the wrong date. |
| Removal | Track each failed cup to its physical disposition | A lane alarm alone does not isolate an individual cup. |
Challenge a defect in each lane and repeat after a recipe or foil change. Include missing results and a failed reject in the line acceptance test.
// FAQ
Questions about yoghurt cup inspection
What does yoghurt cup lid inspection check?
Yoghurt cup inspection checks every cup after sealing: that a foil lid is on and lying flat, and that the best-before date and batch code are printed, readable and correct for the run. It works lane by lane on the filler outfeed and sends a reject signal for each failed cup to the filler.
Can a camera detect a leaking or badly sealed foil lid?
No. A camera sees whether the foil lid is there and whether it lies flat on the rim. A lid can look perfect and still be weakly sealed, and a pinhole or channel in the seal is not visible from above. Seal integrity stays with the sealer’s temperature and pressure controls and your QA leak tests. This is lid inspection, not seal testing.
How much does yoghurt cup inspection cost in South Africa?
An indicative installed budget is R 110 000 – R 680 000 excluding VAT, at 2 October 2026 prices. The low figure is one HIKROBOT vision sensor over a narrow filler; the high figure is two cameras on a vision controller for a wide multi-lane filler, with lid seating checked from the side and an allowance for washdown protection. Both include engineering, installation, FAT and SAT in Gauteng. This is a typical application: every project involves custom requirements, so the final price depends on your products, line and acceptance criteria.
Will the cameras survive washdown in a dairy?
With care. HIKROBOT rates its vision sensors and some cameras IP67 only with the lens cap or lens properly fitted, and lists no stainless washdown housing, so housings and cable glands are part of the station. Cameras are mounted clear of the spray, and the cleaning routine is agreed so hoses are not aimed at them.
Can the same system check date codes on milk bottles?
Yes, as its own station. A milk bottle carries the date on the cap or the bottle, and it is read by OCR and compared with the run in the same way. HIKROBOT shows a fresh-milk case on its Chinese site in which deep learning finds broken, distorted and smeared inkjet dates. The bottle line gets its own camera, recipe and budget.
Does it work on maas and other dairy packaging as well as yoghurt?
The check is the same on any sealed cup or tub: lid on, lid flat, code readable and correct. Maas, yoghurt and other dairy packaging differ in cup size, lid print and where the code sits, so each gets its own recipe, taught from your samples.
Sources
- Milk SA Annual Report 2024 (registered processors; levy product categories) (industry body report, checked 2026-10-02)
- Dairy and imitation dairy products regulations (R260, 2015), regs 25 and 29 (gazette (FAO copy), checked 2026-10-02)
- Regulations governing general hygiene requirements for food premises (R638, 2018) (gazette, checked 2026-10-02)
- HIKROBOT food and medicine cases: fresh-milk inkjet date-code defect detection (Chinese site; figures vendor-stated) (vendor case list (Chinese site, case 鲜奶字符缺陷检测), checked 2026-10-03)
- HIKROBOT Machine Vision Smart Product Catalog 2025Q4: VisionMaster OCV and defect tools (vendor catalogue, checked 2026-10-02)
- HIKROBOT Light Source Catalog 2025Q3: MV-LBES-H flat dome lights (vendor catalogue, checked 2026-10-02)
- HIKROBOT MV-SC3013XC SC3000X vision sensor (OCR tools; IP67 with lens cap) (vendor product page, checked 2026-10-02)
Related:
More line solutions in food and beverage:
Returnable crates →Multipacks →Can ends and dents →Juice box straws →Bread bag closures →Meat tray labels →Wine labels →Sachets and pouches →Staple food bags →Egg cartons and trays →Bottled water (PET) →Food tray seal inspection →Food packaging version verification →Bakery product vision inspection →Returnable glass bottle inspection →
// Next step
Send us cups with good and bad lids and codes
Describe the line, the products and the fault. We test on your good and faulty samples and reply with a feasibility note and a budget.