Journal of Integrative Agriculture ›› 2026, Vol. 25 ›› Issue (3): 1223-1242.DOI: 10.1016/j.jia.2025.05.006

• • 上一篇    下一篇

一个新颖的植被遥感精细提取方法——以油菜为例FI-R

  

  • 收稿日期:2025-01-16 修回日期:2025-05-14 接受日期:2025-04-01 出版日期:2026-03-20 发布日期:2026-02-06

Development of the FI-R model, a novel remote sensing method for fine-scale extraction of vegetation, using rapeseed as an example

Sixian Yin1, Taixia Wu1#, Shudong Wang2, Ran Chen1, Yingying Yang1, Hongzhao Tang3#   

  1. 1 School of Earth Sciences and Engineering, Hohai University, Nanjing 211100, China

    2 Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, China

    3 Land Satellite Remote Sensing Application Center, Ministry of Natural Resources, Beijing 100048, China

  • Received:2025-01-16 Revised:2025-05-14 Accepted:2025-04-01 Online:2026-03-20 Published:2026-02-06
  • About author:Sixian Yin, E-mail: yinsx@hhu.edu.cn; #Correspondence Taixia Wu, E-mail: wutx@hhu.edu.cn; Hongzhao Tang, E-mail: tanghz@pku.edu.cn
  • Supported by:

    This research was supported by the National Natural Science Foundation of China (42201339), and the “Science for a Better Development of Inner Mongolia” Program of the Bureau of Science and Technology of the Inner Mongolia Autonomous Region, China (2022EEDSKJXM003).

PDF

摘要：

植被精细地表信息是研究资源分布状况和环境响应动态变化规律的基础，精确提取不同种属作物的分布对于提升农业生产效率和保障粮食安全具有重要意义。传统的植被精细提取往往因为存在大量具有光谱相似性的地物，模型受背景影响严重，难以在大区域应用。而花期作为植物关键生理周期，在开花时间、花朵形态以及冠层光谱表现等方面都具有独特性，是遥感植被精细提取的有效途径。本研究以油菜为例，选择黄度指标(Blue、Green)和高峰指标(Red、Nir、Swir1)，提出了一种基于Landsat OLI影像的开花植被花期精细提取光谱指数模型(FI-R)，利用NDVI简化同谱异物现象背景影响，成功实现在复杂背景条件下，快速、准确地绘制大范围开花植被分布地图。选择全球五个不同背景的油菜种植区为研究区，验证数据集通过GF影像以及美国CDL农田数据集生成，FI-R表现出更好的区分花期油菜和其它植被的能力，所有研究区总体精度大于94%。FI-R也可以应用于其它相似波段设置的多光谱传感器的油菜提取。该方法在其它开花被子植被精细提取上也具有良好的应用潜力。

Abstract:

The fine-scale characterization of vegetation surface information serves as a fundamental basis for studying the spatial distribution of resources and the dynamic patterns of environmental responses.  Accurately extracting the distributions of different crop species is of critical importance for improving agricultural production efficiency and ensuring food security.  Traditional fine-scale vegetation extraction methods often face significant challenges due to the presence of spectrally similar features and the substantial influence of background interference, which limit their applicability across large areas.  As a key phenological stage of angiosperms, flowering is characterized by distinctive flowering times, floral morphology, and canopy spectral signatures, so it is an effective pathway for fine-scale vegetation extraction using remote sensing.  Using rapeseed as an example, this study developed a spectral index model for precise flowering vegetation extraction (FI-R) based on Landsat OLI imagery.  The model integrates a yellowness index (Blue, Green) and a peak index (Red, Nir and SWIR1) while leveraging the NDVI to mitigate background interference from spectrally similar objects.  This approach successfully enables the rapid and accurate large-scale mapping of flowering vegetation under complex background conditions.  The proposed method was tested in five rapeseed cultivation regions worldwide with diverse backgrounds.  Validation datasets were generated using GF imagery and the U.S. CDL dataset.  The FI-R model demonstrated superior capability in distinguishing flowering rapeseed from other vegetation, and achieved overall accuracies exceeding 94% in all study areas.  Furthermore, FI-R is compatible with other multispectral sensors that have similar band configurations, so it is applicable to rapeseed extraction in broader contexts.  The method also shows strong potential for the fine-scale extraction of other types of flowering angiosperm vegetation.

Key words: flowering rapeseed , multispectral imagery , precise classification , spectral indices , FI-R