Winter Rapeseed dataset 2024

This study proposes a remote sensing mapping Framework for Winter Rapeseed that integrates multi-temporal Optical and synthetic aperture radar (SAR) data - WROS (Winter Rapeseed Optical SAR Integration Framework). This framework is based on an in-depth analysis of the phenological characteristics of the rapeseed growing season, constructs a dynamic time window selection mechanism, and combines a multi-dimensional feature collaboration strategy to enhance the response ability to the dynamic changes of phenology. At the same time, it improves the robustness and regional adaptability of the model in complex environments and is suitable for large-scale winter rapeseed remote sensing mapping tasks. To improve the recognition accuracy, the research innovatively designed two feature indices: The optical Winter Crop Response Difference Index based on Sentinel-2 data and the Dynamic Rapeseed Detection Index based on Sentinel-1 data The DRDI effectively highlights the differences in optical and radar temporal responses between rapeseed and non-rapeseed crops.

本研究提出了一种融合多时序光学与合成孔径雷达（Synthetic Aperture Radar，SAR）数据的冬油菜遥感制图框架——冬油菜光学SAR融合框架（Winter Rapeseed Optical SAR Integration Framework，简称WROS）。该框架基于对油菜生育期物候特征的深入分析，构建了动态时间窗口选择机制，并结合多维特征协同策略，以增强对物候动态变化的响应能力。同时，该框架提升了模型在复杂环境下的鲁棒性与区域适应性，适用于大规模冬油菜遥感制图任务。为提升识别精度，本研究创新性设计了两类特征指数：基于哨兵-2号（Sentinel-2）数据的光学冬作物响应差异指数，以及基于哨兵-1号（Sentinel-1）数据的动态油菜检测指数（Dynamic Rapeseed Detection Index，简称DRDI）。该动态油菜检测指数可有效凸显油菜与非油菜作物在光学与雷达时序响应上的差异。

搜集汇总

背景与挑战

背景概述

Winter Rapeseed dataset 2024是一个用于冬季油菜遥感测绘的数据集，包含澳大利亚、中国和乌克兰的油菜种植区域数据。数据集基于创新的WROS框架，整合了多时相光学和合成孔径雷达数据，特别设计了两个特征指数来提高识别精度。