AGRICULTURE
Weed Identification & Classification
Weed Identification & Classification. What it takes to run it.
Precision targeting for automated weeding and herbicide reduction.
// 01 · The question
“Is that plant a weed, and can we hit it without hitting the crop?”
// 02 · Why it is hard
Crop and weed are both green, both at ground level, both moving under the same wind, and the machine has to decide at the speed the implement is travelling.
// 03 · What the labels have to be
Instance segmentation
Decided before the first frame
A sprayer aims at a region, and the saving comes from spraying the weed instead of the square around it. The mask boundary is what separates the two plants when their leaves overlap, which is most of the time.
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
Weed populations are local. A model that is excellent on one region's species mix meets a different mix two counties over, and the failure looks like a bad model rather than a new distribution.
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