Differential Analysis of Multispectral Images for Terrain Identification
2026-07-10 • Robotics
Robotics
AI summaryⓘ
The authors address the problem of robots having trouble understanding terrain when using normal cameras in tricky lighting conditions. They created DRIFT, a simple system that uses multiple kinds of light measurements, including raw colors and special ratios that reduce lighting effects, to better identify surfaces. Their method highlights differences between these measurements to handle noisy or unreliable data. They tested DRIFT on new images of oil spilled on soil taken by a drone and also on water on grass under different lighting and temperature changes, showing it works better than existing methods and can run efficiently on small devices.
terrain understandingmultispectral imagingband ratiosresidual architecturerobot navigationillumination invarianceedge deploymentMicaSense RedEdge-Punmanned aerial vehiclenear-infrared (NIR)
Authors
Omar Kashmar, Hemendra Arya, Fulvio Mastrogiovanni
Abstract
Reliable terrain understanding is a prerequisite for autonomous robot navigation. Yet, the widespread RGB-based perception can fail under low illumination, shadows, and material ambiguities. In this work we propose DRIFT, a lightweight multispectral framework that combines raw spectral bands and illumination-tolerant band-ratio representations through a dual-stream residual architecture and a differential fusion branch. Band ratios attenuate multiplicative acquisition effects (illumination/sensor gains), while the differential fusion explicitly highlights discrepancies between absolute-band and ratio-derived cues, which improves the robustness to noisy or partially unreliable spectral measurements. In the paper (i) we evaluate DRIFT on a new oil-on-soil multispectral dataset acquired using a MicaSense RedEdge-P camera mounted on an Unmanned Aerial Vehicle, and (ii) we provide an additional controlled study on water-on-grass under varying illumination and thermal perturbations (hot/cold water) to analyze NIR-sensitive effects. DRIFT consistently improves over strong baselines, while remaining compatible with edge deployment.