湖北农业科学 ›› 2020, Vol. 59 ›› Issue (2): 157-160.doi: 10.14088/j.cnki.issn0439-8114.2020.02.035

• 信息工程 • 上一篇    下一篇

基于多时相遥感数据的水稻种植面积信息提取

彭乐文, 张亚   

  1. 河海大学地球科学与工程学院,南京 211100
  • 收稿日期:2019-09-20 发布日期:2020-04-24
  • 作者简介:彭乐文(1995-),女,湖南衡阳人,硕士,研究方向为遥感信息工程,(电话)18351938837(电子信箱)Grit_Peng@163.com

Rice planting area information extraction based on multi-temporal remote sensing data

PENG Le-wen, ZHANG Ya   

  1. School of Earth Sciences and Engineering,Hohai University,Nanjing 211100,China
  • Received:2019-09-20 Published:2020-04-24

摘要: 在遥感影像中,植物的含水量、土壤湿度在短波红外波段下表现很敏感,而红光波段和近红外波段对植物覆盖率、植物长势反映很强烈。基于时间差异的决策树水稻提取模型,通过计算水稻生长不同时期的归一化植被指数NDVI和土壤含水量指数LSWI,在江苏省盐城市射阳县开展了水稻种植区提取的相关研究。经过提取的水稻面积和地方统计数据对比表明,该模型能有效区分出水域、玉米和菜地等较易与水稻种植区混淆的地物,面积提取精度达到76.26%。

关键词: 多时相影像, 决策树, 水稻, 信息提取

Abstract: In remote sensing images, the moisture content and soil moisture of plants were sensitive to the short-wave infrared band, while the red band and near-infrared band reflected the coverage rate and growth rate of plants strongly. According to the characteristics of rice growing in moist soil, based on a decision tree rice extraction model with time difference, this paper carried out relevant research on rice growing area extraction in sheyang county, yancheng city, jiangsu province by calculating NDVI and LSWI of normalized vegetation index and soil moisture index at different growth periods. The comparison of the extracted rice area and local statistical data showed that the model can effectively distinguish the water area, corn and vegetable fields that were easily confused with rice, and the precision of area extraction reached 76.26%.

Key words: multi-temporal image, decision tree, rice, information extraction

中图分类号: