AGRICULTURE
Crop Health & Disease Detection
Crop Health & Disease Detection. What it takes to run it.
Early visual identification of pests, fungus, and nutrient deficiencies.
// 01 · The question
“Which blocks are under stress, and is it spreading?”
// 02 · Why it is hard
Early symptoms are exactly what you want to catch and exactly what looks like ordinary variation, sun scald or dust. The distinction is often a few shades of colour under uncontrolled light.
// 03 · What the labels have to be
Instance segmentation
Decided before the first frame
Disease shows up as discoloured regions with ragged edges, and the treatment decision is driven by how much of the block is affected. Percentage of area is the output, so area has to be the label.
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
Field Precision
Annotate crops, equipment, obstacles, and pathways with polygons, polylines, and semantic segmentation. Tag environmental factors like dust, shadows, and occlusions. 80k+ image annotations in production deployments.
LexInsight
Yield Intelligence
Transform plant-level detections into field-level intelligence. Map disease patterns across zones, track crop health progression over time, estimate yield from visual data.
LexAlert
Navigation Alerts
Monitor autonomous navigation models against seasonal drift. Retrain as fields change to keep navigation reliable through the seasons.
// 05 · What breaks it after launch
The crop changes shape, colour and density every week of the season by design. A model built on one growth stage is out of distribution a fortnight later, on the same field with the same disease.
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 agriculture
GPS-Denied Autonomy
The Challenge
Unreliable GPS under tree canopies and low visibility. Models failed in unstructured, obstacle-heavy terrains.
Our Execution
Annotated thousands of images with polygons and polylines. Tagged diverse environmental factors like dust, shadows, and occlusions.
The Result
Enhanced autonomy in GPS-denied environments. Faster deployment of AI-powered navigation.
80k+
Image annotations in production