发表机构
University of Sannio; European Space Agency; Italian Space Agency (ASI); University of Pavia（桑尼奥大学; 欧洲空间局; 意大利航天局; 帕维亚大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对传统卫星浊度监测延迟高的问题，提出轻量级机器学习方法AquaCubeAI，在Φsat-2星上直接估算沿海浊度，实现低延迟监测，并验证了嵌入式部署的可行性。
AI 中文摘要
及时监测沿海水质对环境保护至关重要，然而传统的卫星工作流程依赖下行链路和地面处理，这会引入延迟，从而可能限制对快速演变的浊度事件的响应能力。为解决这一限制，我们提出了AquaCubeAI，一种轻量级机器学习方法，用于从Φsat-2多光谱图像在轨估算沿海水域浊度。通过将推理从地面段转移到卫星上，AquaCubeAI旨在在星载平台严格的计算和带宽约束下，实现更低延迟、更灵敏且更具操作实用性的浊度监测。该模型在模拟的Φsat-2采集数据上进行训练，这些数据与哥白尼海洋服务（CMEMS）高分辨率海洋颜色（HR-OC）浊度产品在空间上对齐，覆盖了四个欧洲海洋宏观区域的选定局部沿海站点。为了在空间相关性存在的情况下对泛化能力进行现实评估，我们采用了一种空间块分割协议，以减轻训练和评估子集之间的数据泄漏。本研究的主要贡献包括：（i）一个可扩展的数据集生成流程，将模拟的Φsat-2多光谱图像块与CMEMS HR-OC浊度标签配对，覆盖选定的欧洲局部沿海站点；（ii）一个紧凑的基于多层感知器（MLP）的浊度回归器，在泄漏感知的地理空间分割下训练，并针对嵌入式约束进行定制；（iii）通过参数共享对密集空间预测进行重新表述，从而实现浊度制图和简单的基于阈值的异常掩码，用于星上决策逻辑。在Intel Myriad视觉处理单元（VPU）上的嵌入式部署进一步证实了低功耗硬件的可行性，并支持从多光谱输入进行低延迟推理。
英文摘要
Timely monitoring of coastal water quality is critical for environmental protection, yet conventional satellite workflows rely on downlink and ground processing, introducing latency that can limit responsiveness to rapidly evolving turbidity events. To address this limitation, we propose AquaCubeAI, a lightweight machine-learning approach for onboard estimation of coastal water turbidity from Φsat-2 multispectral imagery. By shifting inference from the ground segment to the satellite, AquaCubeAI aims to enable lower-latency, more responsive, and more operationally useful turbidity monitoring under the strict compute and bandwidth constraints of spaceborne platforms. The model is trained on simulated Φsat-2 acquisitions spatially aligned with Copernicus Marine Service (CMEMS) High-Resolution Ocean Color (HR-OC) turbidity products over selected localized coastal sites spanning four European marine macro-regions. To provide a realistic evaluation of generalization in the presence of spatial correlation, we adopt a spatial block splitting protocol that mitigates data leakage between training and evaluation subsets. The main contributions of this work are: (i) a scalable dataset generation pipeline pairing simulated Φsat-2 multispectral patches with CMEMS HR-OC turbidity labels across selected localized European coastal sites; (ii) a compact Multi-Layer Perceptron (MLP)-based turbidity regressor trained under a leakage-aware geospatial split and tailored to embedded constraints; and (iii) a reformulation for dense spatial prediction via parameter sharing, enabling turbidity mapping and simple threshold-based anomaly masks for onboard decision logic. Embedded deployment on an Intel Myriad Vision Processing Unit (VPU) further confirms the feasibility of low-power hardware and supports low-latency inference from multispectral inputs.
Comments16 pages, the paper is under review on IEEE JSTARS
