Open-Source Local Edge AI • BioCLIP + GBIF H3 Geo-Prior • WCAG 2.2 AA (9:1)

Edge-iNat: Sonoran Desert Offline Species & Trail Safety AI

100% On-Device Identification for Yuma County, AZ & Imperial Valley, CA (Zero Cloud Upload)

← alanpruitt.com

In Plain English

Before you read the details below, here is the short version: Edge-iNat identifies desert plants, wildlife, and venomous trail hazards directly on your phone or laptop without needing internet service. It combines a local BioCLIP visual embedding vector with a Yuma / Imperial Valley GPS and seasonal probability filter, then generates a ### verified clinical field receipt.

En palabras sencillas: Edge-iNat identifica plantas, fauna y animales venenosos del Desierto de Sonora directamente en tu teléfono o computadora sin usar internet, combinando visión artificial en el dispositivo con el mes y ubicación en Yuma / Valle Imperial.

1. Edge Vision Input & Spatio-Temporal Geo-Prior

Select a Sonoran Desert field specimen preset below—or upload any local photo / camera frame (`224x224` RGB tensor):

Edge Runtime: Initializing WebGPU / Float32 Vector Engine...

2. Real-Time Edge AI Taxonomic & Safety Audit

Loading Taxa Pack...
Kingdom → Phylum → Class → Order → Family → Genus → Species

--

--

Pattern (Diagnostic Field Marks): --

Habitat & Range: --

Rule & Solve (Clinical / Nutritional Field Protocol): --

Top-3 Edge AI Candidates (Vision Cosine × H3 Monthly Prior)

3. Triple-Hash (###) Clinical Field Receipt & Offline Darwin Core Log

### EDGE-INAT VERIFIED FIELD OBSERVATION ###
Offline Saved Sonoran Desert Field Observations (Darwin Core Log)
Timestamp Scientific & Common Name Family Sector & Month Edge Confidence Safety Gate
No offline observations logged yet. Click "Save to Offline Field Log" above.