Healthy
AI-flagged (pre-symptomatic)
Visibly diseased
Click a leaf to scan it
⚠ Couldn't load the 3D engineThree.js failed to load from the CDN. Check your connection and reload.
A 100-leaf canopy grid renders as an instanced 3D field while a simulated infection — fungal blight or viral mosaic — spreads outward from a patient-zero leaf by logistic growth and diffusion to its neighbours. Every leaf carries its own synthetic 8-band reflectance spectrum (450–860 nm) that drifts from the healthy reference curve as infection progresses, mirroring real red-edge and near-infrared stress signatures used in agricultural remote sensing. Click any leaf, or run "Scan all leaves", to classify it with a genuine hyperspectral algorithm — the Spectral Angle Mapper — which measures the angle between the leaf's measured spectrum and reference curves for healthy, fungal and viral tissue. Because the spectral drift begins well before a human eye can see discoloration, the AI regularly flags leaves as infected while they still look green, and the live scoreboard tracks exactly how many of those pre-symptomatic catches it makes against the count a visual scout would see.