arXiv 2604.044452026-04-06cs.LG

TinyNina：一种资源高效的边缘AI框架，用于通过图像内卫星超分辨率实现可持续空气质量监测

TinyNina: A Resource-Efficient Edge-AI Framework for Sustainable Air Quality Monitoring via Intra-Image Satellite Super-Resolution

ADAPT Research Centre, Dublin, Ireland（ADAPT研究中心，都柏林，爱尔兰）；School of Computer Science, Technological University Dublin, Ireland（都柏林科技大学计算机科学学院，爱尔兰）；School of Computer Science, University College Dublin, Ireland（都柏林大学学院计算机科学学院，爱尔兰）；School of Computer Science and Statistics, Trinity College Dublin（都柏林圣三一学院计算机科学与统计学院）

Prasanjit Dey, Zachary Yahn, Bianca Schoen-Phelan, Soumyabrata Dev

备注：: This manuscript is currently under review at IEEE Access

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出TinyNina框架，利用Sentinel-2多光谱层次作为内部训练标签，实现低资源高精度的NO2监测，实验显示其MAE为7.4μg/m³，比EDSR和RCAN模型效率提升95%且速度提升47倍。

摘要：

氮氧化物（NO₂）是主要大气污染物，对呼吸系统疾病和城市气候挑战有显著影响。尽管Sentinel-2卫星提供全球覆盖，但其原生空间分辨率常限制细粒度NO₂评估。为此，我们提出TinyNina，一种资源高效的边缘AI框架，专门用于可持续环境监测。TinyNina实现了一种新的图像内学习范式，利用Sentinel-2的多光谱层次作为内部训练标签，有效消除了对昂贵且常不可用的外部高分辨率参考数据集的依赖。该框架结合波长特定的注意力门和深度可分离卷积，以保留污染物敏感的光谱特征，同时保持仅51K参数的超轻量足迹。实验结果，基于3,276对匹配的卫星-地面站数据，显示TinyNina实现了最先进的平均绝对误差（MAE）为7.4μg/m³。该性能代表了与高容量模型如EDSR和RCAN相比，计算开销减少了95%，推理速度提升了47倍。通过优先考虑任务特定的实用性和架构效率，TinyNina提供了一种可扩展、低延迟的实时空气质量监测解决方案，适用于智能城市基础设施。

英文摘要：

Nitrogen dioxide (NO$_2$) is a primary atmospheric pollutant and a significant contributor to respiratory morbidity and urban climate-related challenges. While satellite platforms like Sentinel-2 provide global coverage, their native spatial resolution often limits the precision required, fine-grained NO$_2$ assessment. To address this, we propose TinyNina, a resource-efficient Edge-AI framework specifically engineered for sustainable environmental monitoring. TinyNina implements a novel intra-image learning paradigm that leverages the multi-spectral hierarchy of Sentinel-2 as internal training labels, effectively eliminating the dependency on costly and often unavailable external high-resolution reference datasets. The framework incorporates wavelength-specific attention gates and depthwise separable convolutions to preserve pollutant-sensitive spectral features while maintaining an ultra-lightweight footprint of only 51K parameters. Experimental results, validated against 3,276 matched satellite-ground station pairs, demonstrate that TinyNina achieves a state-of-the-art Mean Absolute Error (MAE) of 7.4 $μ$g/m$^3$. This performance represents a 95% reduction in computational overhead and 47$\times$ faster inference compared to high-capacity models such as EDSR and RCAN. By prioritizing task-specific utility and architectural efficiency, TinyNina provides a scalable, low-latency solution for real-time air quality monitoring in smart city infrastructures.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。