OIL & GAS
Corrosion & Structural Degradation
Corrosion & Structural Degradation. What it takes to run it.
Track rust spread on pipes and tanks over time.
// 01 · The question
“How far has this spread since the last inspection?”
// 02 · Why it is hard
The boundary is genuinely ambiguous. Two qualified inspectors outline the same patch differently, so the labelling standard has to be written down before it is applied or the model learns the disagreement.
// 03 · What the labels have to be
Instance segmentation
Decided before the first frame
Corrosion is graded on area and depth of coverage, which is what a mask encodes and a box discards. The label has to be the measurement, or every downstream number is a proxy for how close the inspector stood.
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
Precision Safety
Annotate rig images with bounding boxes for machinery, valves, flanges, and personnel with PPE classification. Safety-focused labels tailored to oil & gas operations. 12k+ hazardous-zone annotations in production deployments.
LexInsight
Corrosion Tracker
Link visual detections to asset management systems. Map corrosion to specific valve IDs, track progression across inspection cycles, connect to maintenance scheduling and compliance codes.
LexAlert
Safety Alerts
Monitor deployed safety models against environmental changes - weather, lighting, equipment repositioning. Detect drift before it causes a missed hazard. Route failures back for immediate retraining.
// 05 · What breaks it after launch
Wet steel is darker steel. Rain, salt spray and low sun all shift the colour and texture that corrosion is identified by, so the same asset reads as worse in the morning than in the afternoon.
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 oil & gas
Hazardous Zone Monitoring
The Challenge
Lacked labeled rig imagery, delaying AI model training. Relied on manual monitoring in hazardous zones.
Our Execution
Annotated rig images with bounding boxes for machinery and personnel. Delivered precise, safety-focused labels.
The Result
Faster AI deployment with rapid model iteration. Improved safety through automated detection.
12k+
Precision image annotations delivered