Every meter of the web inspected – across its full width
nScan inspects paper, films, nonwovens, textiles and sheet metal during running production. Line-scan cameras capture the web without gaps, and the AI detects and classifies defects in real time. We design cameras, lighting and resolution around what your material and your defects require.
- Coverage
- 100%of the web surface, during running production
- Sides
- both sidestop and bottom in one system
- Width
- several metersof web width, scalable with more cameras
- Teach-in
- Teach & Gonew materials taught in seconds to minutes
A defect on the web rarely comes alone
Web materials run non-stop through extruders, coating, laminating or weaving machines. When the process has a problem, the defect often runs on for many meters and otherwise only shows up during converting. A random sample sees only a small part of the web – nScan sees every meter.
Without gaps
The full web width and every meter in the machine direction, on both sides of the web if required.
In real time
Defects are reported immediately during running production – so you can act before the next meters turn into scrap.
Traceable
Every defect is logged with its image and defect class. The type of defect shows where in the process the cause lies.
From paper to the finished garment
nScan is not tied to one material. What matters is the inspection task – the system is matched to your material, your line and your defects.
Paper, board and pulp
Paper, wrapping paper, corrugated board, pulp and tissue – including paper with watermarks.
Films
Plastic films, adhesive films, foam films and bubble wrap – transparent, printed or coated.
Nonwovens
Nonwovens, whose irregular fiber structure is not a defect – but real defects are.
Textiles and leather
Fabrics, also with patterns or seams, up to complete garments. Plus leather with its natural surface.
Sheet metal
Sheet metal and metallic surfaces, also glossy ones – with lighting matched to the surface.
Glass
Glass panes and float glass – down to tiny nickel sulfide inclusions, which we found and made visible in a proof of concept.
Patterns, seams and structure are not defects
Fabrics, nonwovens and printed webs have a structure that repeats constantly or varies naturally. Fixed thresholds have to either let defects through or reject good material.
- Learns your good materialThe AI learns from examples what flawless material looks like – with its pattern, seams and natural variation – and reports only real deviations.
- Teach & GoNew materials and articles are taught in seconds to minutes, in the operator interface and without programming.
- ClassifiedDefects are automatically sorted into defect classes, for targeted optimization of your process.
What nScan finds on the web
Anomaly detection also finds complex defect patterns that nobody had to describe as a rule beforehand. Typical defects on web materials are:
Holes and tears
Holes, tears and cuts in film, paper, nonwovens or fabric.
Inclusions
Foreign particles and inclusions in the material – even very small ones, if the resolution is designed for them.
Contamination
Stains, contamination, adhesions and residues on the surface.
Folds and deformations
Folds, creases, dents and other deformations of the web.
Scratches
Scratches and surface damage, for example on films and sheet metal.
Color, print and pattern
Color deviations and gradients, defects in the print and the pattern, markings and codes.
Camera image
AI heatmap
Camera image
AI heatmap
Camera image
AI heatmap
Camera image
AI heatmap
Camera image
AI heatmap
Camera image
AI heatmapLine by line to a seamless image
A web has no beginning and no end where you could take a photo. That is why web inspection works with line-scan cameras.
Line-scan camera
A line-scan camera captures a single image line across the web. As the web runs beneath it, these lines form a continuous image – with no gaps between the shots.
In step with the web
Image capture runs in sync with the web feed. This keeps the image true to scale, even when the line speeds up or slows down. Several cameras side by side cover wide webs.
Light matched to the material
Front light for the surface, backlight for what is inside the material, or structured light – whatever makes your defects visible.
AI right at the line
The analysis runs directly at the machine, without a cloud. The system processes raw data in the Gbit/s range in real time, thanks to zero-copy from the camera into GPU memory.
React the way you want
Defects go to the line control and the operator interface in real time. You decide what happens next: a defect report per roll, marking on the web or a line stop.
Watermarks – a topic of their own
A watermark is an intended structure in the paper. nScan automatically checks whether it is complete, whether it has defects and how it is embossed.
- Learns by itself, in secondsThe AI watches the running production and learns by itself within seconds what the watermark should look like.
- No training, no golden sampleNo model has to be trained in advance and no golden sample has to be stored.
- Completeness, defects, embossingIncomplete or defective watermarks and deviations in the embossing are reported.
How small can a defect be?
You decide. Web width, speed and smallest defect size determine the choice of camera, optics, lighting and computing power – we build the system that meets your requirement.
A proof of concept shows how far this can go: in it, we found and made visible tiny nickel sulfide inclusions in glass panes.
Built for your web, not for a datasheet
Our platform has a modular design. That is why we use exactly the sensors your task requires – and develop mechanics, electronics, operating system and AI ourselves.
Cameras from any manufacturer
Line-scan or area-scan cameras, monochrome or color – we install the camera that fits your task.
Laser profile sensors
For defects defined by their height, laser profile sensors add a height profile to the camera image.
Reaction to your specifications
A defect report per roll, marking on the web with the marking system of your choice, or a line stop – you decide.
Retrofittable
Integration into existing lines, connected via PROFIBUS, PROFINET, OPC UA or REST API.
Data stays in your plant
Analysis takes place at the line, on our own Wahtari OS instead of a Windows PC.
Ready 24/7
Built for continuous operation, with minimal maintenance and cleaning effort.
Frequently asked questions
Web inspection
Which materials does nScan inspect in web inspection?
Paper, board, corrugated board, pulp and tissue, films, nonwovens, textiles and leather, sheet metal and metallic surfaces, and glass panes. For textiles, also fabrics with patterns or seams and complete garments. What matters is the inspection task – get in touch with us.
How fast can the web run, and how small can defects be?
We design that around your requirement. Cameras, optics, lighting and computing power are matched to web width, speed and smallest defect size. A proof of concept shows how far this can go: in it, we found tiny nickel sulfide inclusions in glass panes.
Why does web inspection use line-scan cameras?
A line-scan camera captures one image line across the web. As the web runs, these lines form a continuous image, with no gaps between the shots. Several cameras side by side cover wide webs, and both sides of the web can be inspected if required.
Which cameras and sensors do you use?
The ones your task requires: line-scan or area-scan cameras from any manufacturer, complemented by laser profile sensors for defects defined by their height if needed. The modular design of our platform makes this possible.
Why AI instead of conventional thresholds?
Fabrics, nonwovens and printed webs have patterns, seams and natural variation. Fixed thresholds have to either let defects through or reject good material there. The AI learns from examples what your good material looks like and reports only real deviations.
What happens when a defect is found?
You decide: a defect report per roll, marking of the defect on the web with a marking system of your choice, or a line stop. Defects are reported in real time to the line control and the operator interface, connected via PROFIBUS, PROFINET, OPC UA or REST API.
Can nScan inspect watermarks in paper?
Yes, automatically. nScan checks whether the watermark is complete, whether it has defects and how it is embossed. The AI watches the running production and learns by itself within seconds – with no training and no golden sample.
Is image data transferred to a cloud?
No. The AI analysis runs directly at the line, and the data stays in your plant. Remote maintenance via the Wahtari Platform is possible, but only with your approval.
Your material, your defects
Talk to us about your web, your speed and the defects you need to find.