发表机构
University of Genoa; Indian Institute of Technology Bombay（热那亚大学; 印度班加罗尔理工学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究针对自主机器人导航中地形识别问题，提出轻量级多光谱框架DRIFT，通过双流残差架构和差分融合分支结合原始光谱带与带比表示，在多光谱数据集评估中优于基线，提升了对复杂情况的鲁棒性且兼容边缘部署。
AI 中文摘要
可靠的地形理解是自主机器人导航的前提。然而，广泛使用的基于RGB的感知在低光照、阴影和材质模糊情况下可能失效。本文提出DRIFT，一个轻量级多光谱框架，通过双流残差架构和差分融合分支结合原始光谱带和耐光照带比表示。带比减弱乘法采集效应，差分融合突出绝对带和比率衍生线索间差异，提高对噪声或部分不可靠光谱测量的鲁棒性。我们在新的土壤上油多光谱数据集和不同光照及热扰动下的草上水数据集上评估DRIFT，它持续优于强基线且与边缘部署兼容。
英文摘要
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.
Comments7 pages, IEEE AIM Conference, 8 Figures
