arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.09175cs.CVcs.LG
收藏

基于热位置编码的卫星图像时间序列广义分类

Generalized Classification of Satellite Image Time Series with Thermal Positional Encoding

Aarhus University(奥胡斯大学);FieldSense A/S(FieldSense公司);IRISA UMR 6074(法国国家信息与自动化研究所雷恩分院(IRISA UMR 6074));Univ. Bretagne Sud(南布列塔尼大学)

Joachim Nyborg, Charlotte Pelletier, Ira Assent

摘要:

大尺度作物类型分类是遥感工作的一项核心任务,其应用具有经济和生态两方面的重要性。当前最先进的深度学习方法是基于自我注意力,并使用卫星图像时间序列(SITS)来基于其独特的生长模式区分作物类型。然而,现有的方法推广到训练过程中没有看到的区域,主要是由于对气候变化引起的生长季节的时间变化不鲁棒。为此,我们提出了热位置编码(TPE)的注意力为基础的作物分类。与之前基于日历时间(例如,一年中的天数)的位置编码不同,TPE基于热时间,热时间通过累积生长季节的日平均温度获得。由于作物生长与热时间直接相关,而不是日历时间,TPE解决了不同地区之间的时间变化,以提高泛化能力。我们提出了多种TPE策略,包括可学习的方法,以进一步改善结果相比,常见的固定位置编码。我们展示了我们的方法在四个不同的欧洲地区的作物分类任务,在那里我们获得了最先进的泛化结果。

英文摘要:

Large-scale crop type classification is a task at the core of remote sensing efforts with applications of both economic and ecological importance. Current state-of-the-art deep learning methods are based on self-attention and use satellite image time series (SITS) to discriminate crop types based on their unique growth patterns. However, existing methods generalize poorly to regions not seen during training mainly due to not being robust to temporal shifts of the growing season caused by variations in climate. To this end, we propose Thermal Positional Encoding (TPE) for attention-based crop classifiers. Unlike previous positional encoding based on calendar time (e.g. day-of-year), TPE is based on thermal time, which is obtained by accumulating daily average temperatures over the growing season. Since crop growth is directly related to thermal time, but not calendar time, TPE addresses the temporal shifts between different regions to improve generalization. We propose multiple TPE strategies, including learnable methods, to further improve results compared to the common fixed positional encodings. We demonstrate our approach on a crop classification task across four different European regions, where we obtain state-of-the-art generalization results.

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。

↑