Existing crop mapping models, rely heavily on reference (calibration) data obtained from remote sensing observations. However, the transferability of such models in space and time, without the need for additional extensive datasets remains a significant challenge. There is still a large gap in developing generalized classification models capable of mapping specific or multiple crops with minimal calibration data. In this study, we present a generalized automatic approach for sunflower mapping at 20-m spatial resolution, using the C-band Sentinel-1 (S1) synthetic aperture radar (SAR) data driven by previously developed phenological metrics. These metrics characterize the directional behavior of the sunflower head, capturing distinct backscattering responses in SAR data acquired from ascending and descending orbits. Specifically, we utilize SAR-derived backscatter values in VH and VV polarization, as well as their ratio VH/VV, as input features to a random forest classifier that was calibrated for the year 2022 in Ukraine. This model is further directly applied to selected sites for multiple years in Ukraine (generalization in time) and other major sunflower producing countries (generalization in space): Ukraine for 2018–2020, and Hungary, France, Russia and USA for 2018. Our results reveal that the model based on features acquired from descending orbits outperforms its ascending orbit counterpart because of the directional behavior of sunflower: user's accuracy (UA) of 96%, producer's accuracy (PA) of 97% and F-score of 97% (descending) compared to UA of 90%, PA of 89% and F-score of 90% (ascending). When generalized to other years and countries, our model achieves an F-score exceeding 77% for all cases, with the highest F-scores (>91%) observed in Ukraine and Russia sites and the lowest (77%) for the US site. We further utilize the produced maps (pixel-based) for the selected regions and years to estimate sunflower planted areas using a statistical sampling-based approach. Our estimates yield the relative root mean square error (RMSE) as 19.7% of the mean area, when compared to the reference data from official statistics and reference maps. These findings demonstrate the robustness of our proposed approach across space and time in generating accurate sunflower maps, its ability to mitigate cloud cover issues through spaceborne SAR data acquisitions, and its potential for obtaining estimates of sunflower planted areas. This research emphasizes the importance of developing interpretable and domain-specific machine learning models that can be readily extended to multiple geographical regions with little to no labelled datasets.

中文翻译：

---

使用 Sentinel-1 SAR 数据绘制向日葵区域的通用模型

现有的农作物绘图模型严重依赖于从遥感观测中获得的参考（校准）数据。然而，此类模型在空间和时间上的可转移性而不需要额外的广泛数据集仍然是一个重大挑战。在开发能够用最少的校准数据映射特定或多种作物的通用分类模型方面仍然存在很大差距。在这项研究中，我们提出了一种通用的自动方法，使用由先前开发的物候指标驱动的 C 波段 Sentinel-1 (S1) 合成孔径雷达 (SAR) 数据，以 20 米空间分辨率进行向日葵绘图。这些指标描述了向日葵头的方向行为，捕获从上升和下降轨道获取的 SAR 数据中的不同反向散射响应。具体来说，我们利用 SAR 导出的 VH 和 VV 偏振反向散射值及其 VH/VV 比值，作为随机森林分类器的输入特征，该分类器是在乌克兰针对 2022 年进行校准的。该模型进一步直接应用于乌克兰（时间上的概括）和其他主要向日葵生产国（空间上的概括）多年的选定地点：2018-2020年乌克兰，2018年匈牙利、法国、俄罗斯和美国。我们的结果研究表明，由于向日葵的方向行为，基于从下降轨道获取的特征的模型优于其上升轨道模型：用户准确度 (UA) 为 96%，生产者准确度 (PA) 为 97%，F 分数为 97%（下降）相比之下，UA 为 90%，PA 为 89%，F 分数为 90%（升序）。当推广到其他年份和国家时，我们的模型在所有情况下都实现了超过 77% 的 F 分数，其中乌克兰和俄罗斯站点的 F 分数最高 (>91%)，美国站点的 F 分数最低 (77%) 。我们进一步利用所选地区和年份生成的地图（基于像素），使用基于统计抽样的方法来估计向日葵种植面积。与官方统计数据和参考地图的参考数据相比，我们的估计得出相对均方根误差 (RMSE) 为平均面积的 19.7%。这些发现证明了我们提出的跨空间和时间方法在生成准确的向日葵地图方面的稳健性，其通过星载 SAR 数据采集减轻云层覆盖问题的能力，以及获得向日葵种植面积估计的潜力。这项研究强调了开发可解释的、特定领域的机器学习模型的重要性，这些模型可以很容易地扩展到多个地理区域，而几乎没有标记的数据集。