一项新的研究论文介绍了一种增强的图像模拟器，用于生成受大气湍流和卫星指向误差影响而退化的真实地球观测（EO）图像。该研究使用此模拟数据评估了YOLOv8和RetinaNet模型在船只检测任务上的性能。结果表明，在退化条件下，YOLOv8的召回率显著下降，而RetinaNet表现出更强的鲁棒性，保持了更高的召回率。 AI
影响 强调了需要更强大的AI模型，在真实的运行条件下进行训练，以实现可靠的地球观测应用。
排序理由 该集群包含一篇学术论文，详细介绍了一种新的模拟方法并评估了现有的AI模型。
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
arXiv:2605.22268v1 Announce Type: cross Abstract: Earth Observation (EO) imagery is often degraded by atmospheric turbulence and pointing jitter; yet, these effects are rarely considered in datasets used to train AI-based detection models. Based on prior work, this paper presents…
Earth Observation (EO) imagery is often degraded by atmospheric turbulence and pointing jitter; yet, these effects are rarely considered in datasets used to train AI-based detection models. Based on prior work, this paper presents an enhanced image simulator that enables the inco…