MANUFACTURING
Weld Quality Inspection
Weld Quality Inspection. What it takes to run it.
Automated analysis of weld porosity, undercut, and spatter.
// 01 · The question
“Is this weld sound, and if not, which defect is it?”
// 02 · Why it is hard
The defect classes look similar to a non-specialist and the acceptance criteria are code-driven, so the labelling standard has to come from the customer's own procedure rather than from general practice.
// 03 · What the labels have to be
Instance segmentation
Decided before the first frame
Weld standards are written in terms of length and area of the discontinuity. Porosity, undercut and spatter are graded by extent, so the label has to be measurable against the same standard the inspector uses.
This is the decision that is expensive to reverse. The geometry has to match what the answer contains, and finding out it does not means labelling the set a second time.
// 04 · How we run it
Three parts of one loop, on this job.
LexAnnotate
Defect Annotation
Pixel-level defect annotation for surface scratches, cracks, dents, weld faults, and assembly misalignments. Multi-sensor alignment for multi-camera inspection systems.
LexInsight
Line Performance
Connect defect detections to production data - batch numbers, machine IDs, shift schedules. Identify defect patterns across lines and predict maintenance needs before unplanned downtime.
LexAlert
Quality Alerts
Monitor defect detection accuracy across changing production conditions. Detect when lighting shifts, new materials, or equipment aging degrades model performance. Auto-retrain to maintain 99%+ accuracy.
// 05 · What breaks it after launch
A change of wire, shielding gas or torch changes the bead's appearance across the board. Every weld now looks slightly unfamiliar, and the model's confidence sags on good and bad alike.
It is not the only one that can get this use case, it is the one that usually gets it first. All ten conditions.
// 06 · In manufacturing
Defect Detection at Scale
The Challenge
Vision models performed well in pilots but lost accuracy as factory conditions shifted.
Our Execution
Built pixel-level annotated datasets and multi-sensor alignment for reliable defect detection.
The Result
Maintained 99%+ defect detection accuracy across changing production environments. Reduced unplanned downtime.
99%+
Accuracy maintained in production