Automated fabric inspection uses line-scan cameras mounted across a fabric inspection frame to find, classify and map every defect in a roll as it runs, then grades the roll to the 4-point system. Inspectors stop searching and start confirming.
Fabric inspection is the check every roll gets before it is cut: an inspector watches fabric run over a lit inspection table, stops at each defect, marks it and records penalty points. It decides whether a roll is accepted, downgraded or returned to the mill.
Automated fabric inspection puts cameras on that same frame. Every metre of the full width is imaged, defects are detected and classified by software, and each one is logged with its position. The result is a consistent grade for every roll and a record you can send back to the supplier.
Cameras mount on an existing inspection frame or a new one. An encoder tracks fabric speed and length so every image is tied to a position in the roll.
Line-scan cameras build a continuous image across the whole width. Reflected light shows surface faults; transmitted light from below shows holes, thin places and weave faults.
Software finds anything that differs from the fabric around it and classifies it: hole, stain, slub, broken end, dropped stitch and so on. Deep learning handles patterned and textured fabrics.
Each defect is scored by size under the 4-point system or your buyer’s rules, and the roll is graded as it finishes.
The operator confirms flagged defects on screen. A defect map, roll report and defect images are saved and can drive cut optimisation.
A line-scan camera captures one line of pixels at a time, so a moving web of any length becomes one continuous image with no gaps or overlaps.
Top lighting reveals stains, oil spots and print faults. Back lighting through the fabric reveals holes, thin places, broken ends and missing picks.
Trained on examples of your own fabrics and defects, a neural network separates real defects from normal texture, slub yarn effects and patterns.
A wheel encoder on the fabric keeps image scale constant as speed changes and gives every defect an exact metre position in the roll.
Two example project scopes showing how the station, the checks and the outputs are defined. Every plant is different, so the final configuration is confirmed after a feasibility study on your own parts.
The situation: A knitting plant inspects every roll by hand and wants consistent grading without replacing its frames.
The situation: A garment plant receives woven rolls from several mills and needs evidence when it rejects one.
Industrial cameras, lenses, line-scan cameras and code readers are available through our machine vision catalog, quoted with delivery to Sri Lanka. We select the hardware for your application during the feasibility study.
Browse the hardware catalogEvery product, surface and line behaves differently under a camera, so every project starts with evidence from your own material.
A site visit to understand your product, line speed, quality criteria and where the check should sit.
We image your good and defective samples and show what the system can reliably detect. This is the go or no-go.
One station on one line, running beside your inspectors until results match and your team trusts them.
More lines, integration with your PLC and systems, and local support from our Sri Lankan engineering team.
Automated surface inspection for scratches, dents, cracks, porosity, stains and coating faults on metal, plastic, rubber and glass. Specialised lighting and deep learning find cosmetic and functional defects on every part.
Industrial code reading and traceability: read and grade barcodes, QR and Data Matrix codes to ISO/IEC 15415 and 15416, verify printed date and lot codes with OCR, and reject every unreadable or wrong code.
Sort fabric rolls, cut panels and washed garments into buyer shade bands with CIELAB measurement and 555 codes.
Every machine vision application we design and build, in one place.
The 4-point system, standardised as ASTM D5430, assigns penalty points to each defect by its length: up to 3 inches is 1 point, over 3 and up to 6 inches is 2, over 6 and up to 9 inches is 3, and over 9 inches or any hole is 4. Points are totalled per roll and normalised per 100 square yards, and the buyer sets the acceptance limit.
Yes, with deep learning. Plain fabrics can be inspected with rule-based methods, but patterns, prints and textured yarns need a model trained on your fabric so that the pattern itself is not flagged as a defect.
It changes their job. The system scans every metre and flags candidates; the inspector confirms or rejects each flag on screen. That keeps human judgement on the decisions that matter while removing the fatigue of watching fabric for hours.
Usually, yes. Most inspection frames can take a camera bar, lighting and an encoder. We check the frame, fabric widths and speeds during a site visit before recommending a retrofit or a new frame.
Yes. The software keeps a profile per fabric type with its own lighting settings, defect classes and grading rules, and the operator selects the profile for each roll.
Tell us what you need to check and share some good and faulty samples. We will show you what a camera can reliably see before you commit to anything.
Photographs show representative plants and equipment, not Cerox installations or client sites. Diagrams by Cerox Engineering. Photo credits, via Wikimedia Commons: Fabric inspection machine by Jamain, CC BY-SA 4.0. Fabric rolls photo by Fahad Faisal, CC BY-SA 4.0.