Textile defect mapping across the overlap of two line-scan views
When wide fabric is inspected by two line-scan cameras, a defect at the view boundary may split, be counted twice or lose its location. Image rows must align with web travel to form a common defect map.
Proposed output: one defect map in fabric-width and travel coordinates, overlap evidence and missing-scan intervals. Registration is a design option, not a throughput or detection-performance commitment.
- 01 Presentation
- 02 Trigger and reference
- 03 Image / profile
- 04 Decision
- 05 Evidence and PLC
Calibrate both views to a reference
- Sample record
- —
- Registration and row validity
- —
- Acquisition progress
- —
Scenes are illustrative simulations; they are not customer references or performance guarantees.
Solution approach
Join backlit image rows by encoder position and calibrate the view overlap against a common reference. Transform AI defect masks into fabric coordinates and merge findings crossing the boundary. Treat missing rows or invalid registration as incomplete inspection. Validate contrast and defect scope with samples.
Process steps
- 01
Presentation
Calibrate both views to a reference
- 02
Trigger and reference
Acquire rows by encoder position
- 03
Image / profile
Extract defect masks
- 04
Decision
Merge boundary findings in one map
- 05
Evidence and PLC
Show incomplete inspection intervals
Related applications
Let us assess your application together
We test your sample in our laboratory and report on resolution, optics and lighting selection.