Machine Perception and Cognitive Robotics Laboratory

AgriSpectraVision

AgriSpectraVision uses AI to simulate hyperspectral imaging from standard RGB cameras, augmented with real soil moisture, temperature, and humidity sensors. Currently monitoring live tomato plants in the lab with cameras and sensors already running — bringing precision agriculture to consumer hardware.

Model Card

RGB camera → AI model simulates hyperspectral bands (NIR, SWIR). Augmented with real sensor array: soil moisture, temperature, humidity. Edge inference on embedded hardware.
Lab tomato plant imagery with paired sensor readings. RGB images annotated with ground-truth spectral data.
Successfully simulates hyperspectral information from consumer RGB cameras. Live monitoring system running in lab.
Spectral simulation accuracy degrades under variable lighting. Currently validated on tomatoes only — needs broader crop testing.

Team

William Edward Hahn, Ph.D.

Founder and Co-Principal Investigator (Co-PI)
Associate Professor

[email protected]

Machine Perception and Cognitive Robotics Laboratory