一种抗时序数据噪声的冬小麦识别方法研究
发布时间：2018-12-12 23:07
【摘要】：冬小麦是全球主要粮食作物之一,及时、准确地获取冬小麦的空间分布信息对于开展冬小麦长势监测、辅助宏观经济决策和应对全球气候变化下的粮食安全等方面具有重要意义。利用2014年9月30日至2015年6月26日冬小麦生长季内35个时相的MODISQ1数据产品,生成NDVI时间序列数据集,在物候分析的基础上选择识别特征,由此构建了一种抗时序数据噪声的冬小麦识别方法,将该方法应用于河南省的冬小麦提取,结果表明,整个研究区的冬小麦总体识别精度达到了93.0%,耕地地块规则区和破碎区的冬小麦识别精度分别为94.0%和86.0%;与传统的基于时间序列提取冬小麦的方法相比,新方法在不同滤波情景下的冬小麦识别精度均较高且稳定,说明该方法可用于基于植被指数时序数据的大范围冬小麦空间分布信息的快速提取,同时也为冬小麦提取工作的业务化实施提供了新的技术支撑。
[Abstract]:Winter wheat is one of the main food crops in the world. Timely and accurately obtaining the spatial distribution information of winter wheat can be used to monitor the growth of winter wheat. Supporting macroeconomic decision-making and addressing food security in the context of global climate change are important. From September 30, 2014 to June 26, 2015, 35 MODISQ1 data products of winter wheat growing season were used to generate NDVI time series data set, and to select identification features based on phenological analysis. A method of winter wheat recognition against noise of time series data was constructed and applied to the extraction of winter wheat in Henan Province. The results showed that the overall recognition accuracy of winter wheat in the whole research area was 93.0. The recognition accuracy of winter wheat was 94.0% and 86.0% in regular area and broken area of cultivated land, respectively. Compared with the traditional method of extracting winter wheat based on time series, the new method is more accurate and stable in different filtering scenarios. This method can be used to extract the spatial distribution information of winter wheat based on vegetation index time series data, and it also provides a new technical support for the operational implementation of winter wheat extraction.
【作者单位】： 北京师范大学地表过程与资源生态国家重点实验室;北京师范大学地理科学学部遥感科学与工程研究院;北京师范大学系统科学学院;中国科学院大气物理研究所大气边界层物理和大气化学国家重点实验室;
【基金】：国家重点基础研究发展计划项目(2015CB953603) 国家自然科学基金项目(41371389) 地表过程与资源生态国家重点实验室资助项目(2017-FX-01(1))
【分类号】：S512.11;TP751
本文编号：2375402
[Abstract]:Winter wheat is one of the main food crops in the world. Timely and accurately obtaining the spatial distribution information of winter wheat can be used to monitor the growth of winter wheat. Supporting macroeconomic decision-making and addressing food security in the context of global climate change are important. From September 30, 2014 to June 26, 2015, 35 MODISQ1 data products of winter wheat growing season were used to generate NDVI time series data set, and to select identification features based on phenological analysis. A method of winter wheat recognition against noise of time series data was constructed and applied to the extraction of winter wheat in Henan Province. The results showed that the overall recognition accuracy of winter wheat in the whole research area was 93.0. The recognition accuracy of winter wheat was 94.0% and 86.0% in regular area and broken area of cultivated land, respectively. Compared with the traditional method of extracting winter wheat based on time series, the new method is more accurate and stable in different filtering scenarios. This method can be used to extract the spatial distribution information of winter wheat based on vegetation index time series data, and it also provides a new technical support for the operational implementation of winter wheat extraction.
【作者单位】： 北京师范大学地表过程与资源生态国家重点实验室;北京师范大学地理科学学部遥感科学与工程研究院;北京师范大学系统科学学院;中国科学院大气物理研究所大气边界层物理和大气化学国家重点实验室;
【基金】：国家重点基础研究发展计划项目(2015CB953603) 国家自然科学基金项目(41371389) 地表过程与资源生态国家重点实验室资助项目(2017-FX-01(1))
【分类号】：S512.11;TP751
【相似文献】
相关期刊论文 前5条
1 郝善勇,刘玉树;基于样本数据重抽样的时序数据预报方法[J];北京理工大学学报;2000年05期
2 贾明明;刘殿伟;宋开山;王宗明;姜广甲;杜嘉;曾丽红;;基于MODIS时序数据的澳大利亚土地利用/覆被分类与验证[J];遥感技术与应用;2010年03期
3 董辉;傅鹤林;冷伍明;;滑坡位移时序预测的核函数构造[J];岩土力学;2008年04期
4 王坚,孙祥中,高井祥;形变时序数据性态特征确定方法研究[J];测绘工程;2004年01期
5 李文梅;覃志豪;李文娟;裴欢;;基于MODIS-NDVI时序数据的南方冬闲田信息提取[J];中国农学通报;2010年07期
，本文编号：2375402
本文链接：https://www.wllwen.com/guanlilunwen/gongchengguanli/2375402.html
