基于MODIS-EVI时间序列与物候特征的水稻面积提取
作者:
作者单位:

作者简介:

通讯作者:

中图分类号:

基金项目:

江苏省农业科技自主创新资金项目(CX(20)3153)和江苏省自然科学基金项目(BK20200281)


Extraction of Rice Planting Area Based on MODIS-EVI Time Series and Phenological Characteristics
Author:
Affiliation:

Fund Project:

  • 摘要
  • |
  • 图/表
  • |
  • 访问统计
  • |
  • 参考文献
  • |
  • 相似文献
  • |
  • 引证文献
  • |
  • 资源附件
  • |
  • 文章评论
    摘要:

    物候是植被生理生态过程与环境变化相互作用的体现,时间序列遥感数据的使用有助于揭示水稻物候特征。基于水稻物候特征建立一个可靠的水稻面积监测体系,及时、准确地监测水稻种植面积,对于粮食安全十分重要。本研究以中等分辨率成像光谱仪(Moderate resolution imaging spectroradiometer, MODIS)为数据源,选择增强型植被指数(Enhanced vegetation index, EVI),重构2019年和2020年EVI时间序列,提取水稻物候信息,并选择季节积分和生长季振幅两个指标,结合2019年单点EVI时间序列和水稻种植面积的统计数据,确定江苏省13个地级市水稻的季节积分和生长季振幅的阈值,并根据得到的阈值,提取2020年江苏省水稻种植面积。利用2020年水稻种植面积的统计数据和美国陆地卫星-8携带的陆地成像仪(Landsat8 operational land image, Landsat8 OLI)影像,对提取结果进行了精度验证。结果表明,水稻提取的总体精度为92.55%,Kappa系数为0.8463,水稻的制图精度为92.90%,用户精度为89.09%,与统计数据的一致性为93.90%,提取精度较高,在技术上具有可行性。该方法为大区域提取农作物种植面积提供了参考。

    Abstract:

    Rice is the second largest crop in China, and its planting area and spatial distribution information are the main basis for the adjustment of crop planting structure. Phenology is the reflection of the interaction between vegetation physiological and ecological processes and environmental changes, and the use of time series remote sensing data can help reveal the phenological characteristics of rice. The moderate resolution imaging spectroradiometer (MODIS) was used as the data source, the enhanced vegetation index (EVI) was selected to reconstruct the EVI time series and extract rice phenological information in 2019 and 2020. Area total and amplitude were selected as extraction indicators, combining with the single-point EVI time series in 2019 and the statistical data of rice planting area, the thresholds of area total and amplitude for 13 prefecture-level cities in Jiangsu Province were determined. According to the obtained thresholds, the rice planting area in Jiangsu Province in 2020 was extracted. Finally, the accuracy of the extraction results was verified by using the statistical data of the rice planting area in 2020 and Landsat8 images. The results showed that the overall accuracy of rice extraction was 92.55%, the Kappa coefficient was 0.8463, the mapping accuracy of rice was 92.90%, the user accuracy was 89.09%, and the consistency with statistical data was 93.90%. The research results can provide a reference value for extracting crop planting area in large areas.

    参考文献
    相似文献
    引证文献
引用本文

田苗,单捷,卢必慧,黄晓军.基于MODIS-EVI时间序列与物候特征的水稻面积提取[J].农业机械学报,2022,53(8):196-202. TIAN Miao, SHAN Jie, LU Bihui, HUANG Xiaojun. Extraction of Rice Planting Area Based on MODIS-EVI Time Series and Phenological Characteristics[J]. Transactions of the Chinese Society for Agricultural Machinery,2022,53(8):196-202.

复制
分享
文章指标
  • 点击次数:
  • 下载次数:
  • HTML阅读次数:
  • 引用次数:
历史
  • 收稿日期:2022-04-06
  • 最后修改日期:
  • 录用日期:
  • 在线发布日期: 2022-05-19
  • 出版日期: