引用本文:
【打印本页】   【HTML】   【下载PDF全文】   查看/发表评论  【EndNote】   【RefMan】   【BibTex】
过刊浏览    高级检索
本文已被:浏览 135次   下载 0  
分享到: 微信 更多
近20年华北地区冬小麦面积遥感快速提取研究
黄青1, 杜彦彦2
1.中国农业科学院农业资源与农业区划研究所;2.中国东方红卫星股份有限公司
摘要:
【目的】基于遥感数据,研究快速提取近20年(2001-2020年)华北地区冬小麦种植面积的方法,生产准确的长时间序列冬小麦面积遥感产品,为政府决策部门和科研单位科研工作提供数据支撑。【方法】基于MODIS植被指数产品,经过滤波重构之后,分析华北地区不同纬度冬小麦在整个生长季的的时序特征,考虑不同区域冬小麦物候差异,提出了一种关键生长季时序NDVI曲线匹配的方法,在无样本的条件下,快速提取提取冬小麦面积。使用统计年鉴进行提取面积的验证,同时使用目视解译的样本和高分数据哨兵2号提取的结果,计算混淆矩阵,进行精度评价。【结果】与2001-2018年统计年鉴对比,平均相对误差为16.1%,与目视解译和哨兵2号分类结果中6459个采样点的精度评价,总体精度达87.1%,kappa系数为0.65。【结论】根据冬小麦物候特征,提取NDVI的时序特征,利用时序NDVI曲线匹配算法,可实现华北地区冬小麦面积和种植分布的快速提取。
关键词:  遥感  MODIS  冬小麦  华北地区
DOI:
分类号:
基金项目:“中央级公益性科研院所基本科研业务费专项”(1610132020017)
Extraction of winter wheat in North China in recent 20 years using remote sensing
huangqing1, duyanyan2
1.the Institute of Agricultural Resources and Regional Planning, Chinese Academy of Agricultural Sciences;2.China Spacesat Co., Ltd
Abstract:
[Purpose] To study the rapid extraction method of winter wheat planting distribution in North China from 2001 to 2020, and to produce accurate long time series winter wheat remote sensing products, so as to provide data support for the scientific research of government decision-making departments and research institutions. [Method] Based on MODIS vegetation index product, after filtering and reconstruction, the time sequence characteristics of winter wheat in different latitudes in North China during the whole growing season were analyzed. Considering the difference of winter wheat phenology in different regions, a new method of NDVI curve matching for the time sequence of key growing seasons was proposed to extract winter wheat rapidly without samples. The statistical yearbook was used to verify the extraction area, and the obfuscation matrix was calculated using the samples interpreted visually and the results extracted by Sentry 2 to evaluate the accuracy. [Result] Compared with the statistical yearbook from 2001 to 2018, the mean relative error was 16.1%. Compared with the accuracy evaluation of 6459 sampling points in the visual interpretation and Sentry 2 classification results, the overall accuracy reached 87.1%, and the kappa coefficient was 0.65. [Conclusion] According to the phenological characteristics of winter wheat, the temporal features of NDVI can be extracted, and the area and planting distribution of winter wheat in North China can be quickly extracted by using the temporal NDVI curve matching algorithm.
Key words:  Remote sensing  MODIS  winter wheat  North China