Detection
Every plant, every frame
Detection and instance segmentation of crop and weed plants at implement speed.
Perceive · Folira
Folira is Cropite's edge perception engine. It runs on Jetson, reads multiple cameras on a moving machine and labels every plant: crop or weed, pest damage, disease, nutrient stress and ripeness.
ExampleClasses and confidences are illustrative.
Capabilities
Detection
Detection and instance segmentation of crop and weed plants at implement speed.
Pipelines
DeepStream pipelines with on-device tracking across cameras and frames.
Model packs
Packs tuned to a crop, its weeds and its soils, updated over the air.
Uncertainty
Confidence and novelty scores flag plants it has not seen for labelling.
Output
A structured detection record that Targis, Acreon and Agrona read.
Conditions
Training covers dust, glare, low sun and plants that overlap.
Detection classes
The first model pack is planned for one row crop and its common weeds. Everything else is on the draft roadmap.
| Detection class | Kind | Crops | Model pack |
|---|---|---|---|
| Soybean | Crop | soybean | First pack (planned) |
| Corn | Crop | corn | First pack (planned) |
| Lettuce | Crop | lettuce | Later (planned) |
| Grapevine | Crop | grape | Later (planned) |
| Amaranthus (pigweed) | Weed | soybean, corn | First pack (planned) |
| Grass weeds | Weed | soybean, corn | First pack (planned) |
| Weeds in lettuce beds | Weed | lettuce | Later (planned) |
| Leaf spot | Disease | soybean | Later (planned) |
| Downy mildew | Disease | grape, lettuce | Later (planned) |
| Insect feeding damage | Pest | soybean, corn | Later (planned) |
| Nitrogen stress | Stress | corn | Later (planned) |
| Ripe bunch | Ripeness | grape | Later (planned) |
No class matches both filters. Choose All in one of them.
How it works
1
Cameras on the boom stream frames to the edge unit.
2
TensorRT models find and label every plant.
3
Each plant gets an ID that survives from frame to frame.
4
A detection event goes to Targis in milliseconds.
Specification
| Item | Planned |
|---|---|
| Edge hardware | Jetson Orin or IGX |
| Cameras per unit | Up to 8 (target) |
| Frame to detection | <30 ms (target) |
| Classes | Crop, weed, pest, disease, stress, ripeness |
| Model delivery | Over the air, per machine group |
| Output | JSON events, as-applied maps |
Status
Nothing on this page is shipping today. Items marked planned are on the draft roadmap.
Place in the loop
Folira is where the loop starts. Its detections drive Targis's actions, fill Acreon's twin and tell Agrona where to send the next scouting pass. Plants it is unsure about go to labelling, and Synora makes more examples like them.
Questions
An agronomist or an engineer reads every message, not a sales queue.
Industrial RGB cameras mounted over the rows. The kit and layout depend on the machine; we specify them per setup.
With lighting on the boom, yes. Night passes are planned after daytime packs are proven.
Groups first (broadleaf and grass), then species where it changes the treatment.
Next step
Send us one pass. We run it through Cropite and send back a per-plant report, labelled as an early example.