欢迎访问《中国农学通报》,

中国农学通报 ›› 2019, Vol. 35 ›› Issue (16): 141-147.doi: 10.11924/j.issn.1000-6850.casb18030081

• 工程 机械 水利 装备 • 上一篇    下一篇

基于HJ卫星的中国南方地区甘蔗面积提取研究

张东东1,2,周振2,3,宋晓东2   

  1. 1.西藏高原大气环境科学研究所;2.江苏省无线电科学研究所有限公司;3.浙江大学环境与资源学院
  • 收稿日期:2018-03-15 修回日期:2019-05-10 接受日期:2018-05-25 出版日期:2019-06-04 发布日期:2019-06-04
  • 通讯作者: 宋晓东
  • 基金资助:
    国家公益性行业(气象)科研专项重点项目“甘蔗产量预测及气象灾害监测评估技术研究”(GYHY201406030);国家自然基金项目“基于 MODIS FY-3 VIRR遥感数据的藏北地区干旱遥感监测方法”(41465006)。

Study on Sugarcane Crop Classification in Southern China Based on HJ-1 CCD Images

  • Received:2018-03-15 Revised:2019-05-10 Accepted:2018-05-25 Online:2019-06-04 Published:2019-06-04

摘要: 为了研究大范围甘蔗种植面积的提取方法,以广西、云南、广东湛江和海南为研究区,以30 m空间分辨率的多时相HJ卫星影像为数据源,采用基于NDVI时间序列的决策树分类模型提取研究区内2014/2015年度甘蔗种植面积。结合农业部门的统计数据对甘蔗种植面积提取结果进行精度评价,总体精度达到87.5%。对研究区广东湛江甘蔗种植区域进行抽样调查,抽样调查精度达到93.2%,Kappa系数为0.81。表明该方法可以高效地应用于中国南方地区的甘蔗种植空间信息识别。

关键词: 暗紫贝母, 暗紫贝母, 施肥, 整地, 种植密度, 生长高度, 存活率

Abstract: This study aims to propose an extraction method of sugarcane planting in large area using remote sensing data. Taking Guangxi Zhuang Autonomous Region, Yunnan Province, Hainan Province and Zhanjiang City in Guangdong Province as the study areas, the time-series Chinese HJ- 1 CCD images were obtained covering the sugarcane growing period in the study areas. The decision tree classification method was applied on the base of the time-series NDVI threshold for sugarcane mapping over the large areas during 2014/2015 sugarcane harvest year. The overall accuracy was 87.5% compared with the statistics of local agricultural department. The independent field survey sampling points were used to evaluate the classification accuracy in Zhanjiang City. The confusion matrix analysis showed that the overall classification accuracy was 93.2% and the Kappa coefficient was 0.81. The results showed that this method was feasible, efficient, and applicable in extracting the planting area of sugarcane in southern China.

中图分类号: