Gani
IoT & Computer Vision Early-Warning Platform for Mine Safety
An IoT and machine learning early-warning platform for predicting and detecting rockfalls in open-pit mines. Gani fuses real-time vibration, tilt, and seismic telemetry from ESP32 microcontroller arrays with YOLOv8 visual detection into a Next.js multi-role dashboard with severity-tiered alerts (Low, Medium, Critical).
System Specifications
key technical metricsSystem Architecture & Methodology
engineering breakdownMultimodal Sensor Telemetry
Fuses live seismic vibration, structural tilt, and geological movement readings from ESP32 edge nodes with optical CCTV feeds for high-confidence event detection.
YOLOv8 Rockfall Computer Vision
Runs real-time computer vision inference on hazardous slope surfaces to identify early rock displacement and slope fractures before structural collapse.
Tiered Alert & Monitoring Console
Engineered a low-latency Next.js monitoring dashboard categorizing threats into Low, Medium, and Critical alerts with instant sound triggers and visual beacons.
Resilient Offline Operation
Engineered edge nodes to log sensor telemetry locally and maintain core alert thresholds even in transient connectivity and harsh mine conditions.