Bakery shape and colour inspection
Bakery appearance inspection needs recipe-specific examples of acceptable shape and browning. Distinguish presentation from actual size.
Bakery product vision inspection checks visible shape, size, broken edges, relative surface colour and arrangement before or during packing. A camera compares products with approved examples for the active recipe, and the line isolates agreed defects. Crumbs, overlap, normal baking variation and belt background must be included in the sample test. Surface appearance cannot prove internal bake quality, microbiological safety or the absence of hidden foreign material.
A broken biscuit and a whole biscuit overlapping another can produce similar outlines. The station needs a presentation that exposes the feature used for the decision. Decide whether inspection belongs on separated products before packing, on an open tray, or at both points. Product inspection answers whether the visible item is acceptable; bread-bag closure inspection answers whether its bag has been closed and coded. Neither should be used as evidence for the other.
This is an engineering scope for sample assessment, not a report of a completed customer installation. The sources below support measurement, imaging or test-method principles; station performance must be established on your products.
| What the line reports | What a station checks | Where it stops |
|---|---|---|
| Broken biscuits reach packing | Inspect exposed contours and missing regions | Overlapping products may conceal the broken edge. |
| Surface appearance varies | Compare relative colour/appearance under controlled illumination against the recipe’s accepted range | Colour is not a direct test of internal bake or safety. |
| A tray is missing a product | Check expected occupied positions before lidding | Covered or stacked layers need their own inspection point. |
| Crumbs trigger too many alarms | Validate the background, normal crumbs and real defect examples | Do not simply suppress all small regions if they represent a critical defect. |
Swipe sideways to see the whole drawing →
| Component | HIKROBOT example | Role in this station |
|---|---|---|
| Camera | HIKROBOT area-scan cameras | Image separated products or open pack positions. |
| Lighting | Stable colour or monochrome illumination appropriate to the decision | Keep variation in the lighting from becoming apparent product variation. |
| Presentation | Defined product lanes and trigger/encoder | Keep identity between the captured product and the reject. |
| Environment | Protective mounting compatible with the cleaning plan | Consider dust, crumbs, access and safe cleaning of optics. |
| Tool or software | What it does here |
|---|---|
| Recipe reference | Store agreed acceptable product variations, not one ideal image. |
| Shape and completeness | Locate each visible item and inspect exposed outline/occupied regions. |
| Appearance classification | Use colour or trained classification only for defined visible defects. |
| Reject tracking | Associate each result with its lane, position and reject window. |
Agree the defect/feature list, acceptable variants, reference samples, failure response and excluded checks.
Provide approved appearance ranges for each recipe and normal baking variation. Include actual broken, misshapen and wrongly arranged products plus borderline cases.
Define permitted spacing, lanes and how fragile rejects will be removed. A result is valid only for its own part or pack cycle.
Bind the program and inspection limits to the authorised product revision; record who changes them.
Repeat the challenge set under the agreed production conditions, then check cleaning, restart, bypass and rework handling.
The optical test and integration survey define the scope before a rand budget is issued.
No numerical budget is assigned to this scope before optical feasibility and the handling/PLC requirements are defined. A shared camera model or similar product does not establish the cost of this complete station.
| Cost driver | What changes the scope | Information needed |
|---|---|---|
| Coverage | Number of lanes and whether individual products are separated | Line video/layout and representative samples |
| Defect definition | Shape, visible damage, relative colour or tray completeness | Approved examples and rejection rules |
| Handling and hygiene | Fragile product removal and cleaning access | Site process and cleaning requirements |
Compare the scope with the cost guide. The quotation states its price date, VAT treatment, installation, commissioning and exclusions.
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 |
|---|---|---|---|
| Surface defect detection | Core | Supports the bakery product vision inspection task within the stated viewing and acceptance limits. | Surface defect detection |
| Colour identification | Core | Supports the bakery product vision inspection task within the stated viewing and acceptance limits. | Surface and colour inspection |
| Presence detection and counting | Supporting | Supports the bakery product vision inspection task within the stated viewing and acceptance limits. | Presence and counting |
| PLC link and reject with confirmation | Supporting | Supports the bakery product vision inspection task within the stated viewing and acceptance limits. | PLC and reject integration |
Every station is accepted on your products, against numbers agreed in writing before the build. How we deliver →
Measure false rejects and misses by product family and defect class, not only an overall average.
Include crumbs, belt joints, touching products and the defined overlap limit.
Prove physical removal without losing the identity of adjacent acceptable products.
Record the challenge set, outcomes, recipe revision, inspection conditions and unresolved limits. Agree false-reject and missed-defect targets before building; no universal performance figure is assumed.
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.