Manufacturing & Industry
Plastics
In plastics production most defects appear on transparent or translucent surfaces, a difficult problem class for standard imaging.

Problems we address in this sector
- 01Defect detection on transparent and translucent surfaces
- 02Modelling how injection and extrusion parameters affect quality
- 03Material classification in recycling streams
- 04Objective measurement of colour and tone deviation
Where it gets difficult
In injection moulding and extrusion, many defects form not on the surface but at the edge or in the geometry of the part: flash, short shots, flow lines, sink marks. On transparent parts a defect becomes visible only under special lighting such as backlight or polarised light.
Quality usually depends on machine parameters. Keeping the inspection record together with mould temperature, pressure and cycle time lets you find the cause of a defect, not just sort the part out.
Approaches we apply
Frequently asked questions
Is defect detection possible on transparent parts?
Yes, but not with standard lighting. A setup in which the defect separates in the image (backlight, polarisation or angled light) is tested during feasibility.
Does the model need retraining when the mould changes?
The model is adapted to the new part geometry. This is fine-tuning with a small number of samples; once the process is defined, a mould change is not a long project.
Can it connect to machine data?
Yes. Inspection decisions can be stored in the same record as machine parameters, which shows which parameter change a rise in defect rate started with.
Other sectors

Manufacturing
Visual quality inspection, process optimisation and industrial automation.
Explore
Automotive
Surface defect detection, assembly verification and vision models for driver assistance.
Explore
Chemicals
Production monitoring, quality control and materials discovery.
Explore
Pharmaceuticals
Packaging and serial number verification, retrieval with source attribution and decision trails.
Explore
Food
Production quality control, supply chain and adulteration detection.
Explore
Do you have a problem in this sector?
Let's define the problem together in a technical consultation; feasibility is assessed at the first stage.

