Cosmetics and personal care
Catch fill-level errors, label and pump-cap defects, decoration drift, batch-code legibility, and contamination before product leaves the filling line.
Automated quality inspection for cosmetics and personal-care production, running on a refurbished iPhone alongside your filler, capper, decoration cell, labeller, and case packer.
What is automated quality inspection for cosmetics and personal-care production?
AI defect detection for cosmetics and personal care uses a camera and a vision model to watch every bottle, jar, and tube as it leaves the filler, the capper, the labeller, and the secondary-pack station, and to flag non-conforming units before they reach the depot. Instead of an operator at the inspection table or rigid rule-based vision, the model learns the bottle silhouette, label artwork, decoration finish, and batch-code area of your SKU portfolio, and applies a consistent visual checkpoint across shifts, line speeds, and SKU changeovers.
Cosmetics and personal-care products are particularly hard to inspect at line speed because the bulk colour reads differently across opaque, translucent, and clear formulations, the decoration artwork on a prestige bottle is intentionally subtle, and the underfilled jar that ruins a gift-set looks identical to a normal head-space variation under cleanroom lighting. Rule-based vision built around a single SKU breaks the moment you swap to a different bottle, a different decoration, or a different recipe. AI-led inspection handles those variations because the model learns from real production frames rather than from a fixed threshold.
The result is an automated visual checkpoint that complements your end-of-line sample and gives you a unit-by-unit image record. When a retailer query comes back six weeks later, you can pull the frames from the exact production window and either confirm the defect or push back with evidence.
Defects we catch on cosmetics and personal-care production lines
Fill-level errors and head-space drift
Fill-level errors are the under- and over-filled bottles caused by filler-piston wear, bulk-viscosity changes during the run, or temperature drift in the bulk holding tank. Underfills break the labelled-volume spec at the retailer's depot, and overfills cause cap-thread contamination at the capper. Operators check fill by eye but cannot watch every unit, so the borderline cases pass the inspection table. The AI model learns the in-spec head space for each SKU and flags drift as soon as the local fill height crosses your tolerance, with the frames available so you can adjust the filler before a full pallet ships out of spec.
Pump-cap and closure seating
Closure defects include cocked pumps, missing dip-tubes, off-centre screw caps, and underseated dispenser caps caused by capper torque drift, cap-feed misalignment, or thread tolerance issues. Cocked pumps fail the dispense test on a vanity, and underseated caps leak in transit. Operators sample caps at break but miss the windows in between. The AI model learns the in-spec cap signature and flags cocked, missing-dip-tube, and underseated caps at the capper exit, with the frames available so you can adjust torque before a full batch ships.
Label misalignment and decoration drift
Label and decoration defects include skewed application, lifted corners, misaligned hot-stamp foil, and rotation errors caused by glue-roller wear, label-stack feed errors, decoration-drum drift, or pressure-roller misalignment. The worst offenders sit on the back of a prestige bottle and pass the front-of-label inspection table to fail at the depot. The AI model holds the visual signature of an in-spec label and decoration for each SKU and flags skew, lift, and rotation as soon as the local pattern deviates from spec.
Hot-stamp, screen-print, and foil defects
Decoration-print defects include broken foil, missing screen-print pixels, and feathered hot-stamp text caused by foil-roller wear, screen-mesh contamination, or stamp-temperature drift. The defects fail prestige-listing inspection and trigger consumer complaints from gift-set returns. The AI model learns the in-spec decoration signature for each SKU and flags broken foil, missing pixels, and feathered text at the decoration cell so the line can adjust before a full run ships.
Batch-code legibility and recipe stamps
Batch-code defects include faded ink, smudged digits, missing characters, and wrong recipe codes caused by inkjet printer maintenance, ribbon wear, or recipe-changeover errors. The defects fail retailer goods-in inspection and trigger pallet rejections at the depot. The AI model reads the batch-code area in every frame and flags illegible, missing, or wrong-format codes at the labeller exit so the line can correct the printer before a full pallet ships.
Bottle and tube cosmetic damage
Cosmetic defects include surface scratches on glass, dent-marks on tubes, embossment errors, and shoulder-flow lines caused by transfer-belt wear, capper impact, or supplier batch issues. The worst cases survive the inspection table and fail at the prestige-retailer goods-in. The AI model learns the in-spec cosmetic signature and flags scratches, dents, and embossment errors at the case-packer entry so the line can divert before the case wraps it.
The lighting setup that makes this work on a cosmetics line is a diffuse overhead light over the filler and labeller to read fill level, label, and decoration, plus a low-angle ring light at the capper to read closure seating and a backlight at the case packer to read pack composition. An iPhone Pro with macro and wide-angle lenses handles the seven defect families from a single inspection station per critical control point. We synchronise the rig with the conveyor encoder so that flagged units drive a downstream divert or hold decision. We spec the optics with you during onboarding.
How Enao runs on a cosmetics and personal-care line
The full hardware rig costs less than €1,000 and consists of a refurbished iPhone Pro, a diffuse overhead light with an optional low-angle ring light for capper inspection and a backlight for case-pack composition, a USB-C cable, and a mount that clamps over the filler, the capper, the decoration cell, the labeller, or the case packer. PLC integration is not required for the first deployment, the rig fits in a flight case, and the line keeps running while you set it up.
Onboarding is self-serve. Your line team mounts the rig, opens the Enao app, and starts collecting reference frames at the next changeover. Day one returns 80% accuracy without any prior labelling, and by day fourteen the model is operating above the manual inspector on the defect families it has seen, improving with every flagged unit that the line confirms or rejects.
Each line teaches its own model what its bottle shapes, decoration artwork, and cap geometries look like. When you swap to a different SKU or bulk recipe on the same line, the model adapts in a single shift. When you bring a sister line online with a similar product family, the second model starts from the first model's experience and the marginal effort drops sharply.
Out-of-spec units stop reaching the case packer, scrap is logged at the inspection point rather than at the QC office, and your operators get back the hours of attention they need for the parts of the job that still need a human, including filler setup, recipe tuning, and consumer-return analysis.
How Enao compares to manual inspection and traditional machine vision
For cosmetics and personal-care producers the comparison sharpens around five dimensions.
Setup time on a cosmetics line. — Manual visual inspection: hours of training per operator, ongoing labour. Traditional machine vision (softwebsolutions, scortex, Cognex, Overview.ai, intelgic): three to nine months of integration with a system integrator, plus a rule set per SKU. Enao: deployed in a week by your own team on a refurbished iPhone, day one at 80% accuracy.
Hardware cost per line. — Manual visual inspection: none upfront, ongoing labour cost. Traditional machine vision: €40,000 to €200,000 per line for industrial cameras, structured lighting, and integration. Enao: under €1,000 per line with a refurbished iPhone Pro, lamp, and mount.
Handling new SKUs, decorations, and bulk recipes. — Manual visual inspection: re-train operators for every new SKU. Traditional machine vision: rewrite the rule set per recipe, often outsourced to the integrator. Enao: re-teach the model on new bottles, decorations, and recipes in a single shift, no code to touch.
Detection accuracy on subtle decoration drift and cosmetic damage. — Manual visual inspection: high at shift start, drops measurably after three hours. Traditional machine vision: strong on dimensional checks, weak on subtle decoration drift and cosmetic-damage detection. Enao: learns fill, label, decoration, and cap signatures from reference frames and holds accuracy across shifts and runs.
Who runs it. — Manual visual inspection: trained operator at the inspection table. Traditional machine vision: system integrator or a specialised vision engineer. Enao: your line team, no external specialist required.
Beauty retailers and prestige-distribution category managers change vendors over the cost of a rejected pallet, and the cost of a chargeback or a quiet listing swap sits well above the cost of an iPhone-based inspection rig. Enao is built for that gap.
