摘要
为快速获取大范围种植结构复杂区域的作物种植面积,以MODIS数据为数据源,选择归一化植被指数(Normalized difference vegetation index,NDVI)、增强植被指数(Enhanced vegetation index,EVI)、宽动态植被指数(Wide dynamic range vegetation index,WDRVI)、地表水分指数(Land surface water index,LSWI)、归一化雪被指数(Normalized difference snow index,NDSI)5种特征,结合同步的实地调查样本点,采用支持向量机算法(Support vector machines,SVM)提取黑龙江省主要农作物的种植面积。研究表明,在待选特征中NDVI、EVI与LSWI指数组合取得了最高的分类精度,总体分类精度为74.18%,Kappa系数为0.60;支持向量机算法与最大似然算法、随机森林算法相比,分类精度更优。该方法为在大区域中提取农作物种植面积提供了参考价值。
Mapping the crop planting pattern and cropped area rapidly and accurately in Heilongjiang Province is important for agrieultural monitoring. MOD09 and MOD13 were selected as data source for its high time resolution and good quality. To explore the optimal feature and classification method which can obtain the spatial distribution of the main crops in Heilongjiang Provinee, NDVI, EVI, WDRVI, LSWI and NDSI were selected as input data for crop classification based on time-series of MODIS data and combined with field survey sample points. The results showed that the combination of NDVI, EVI and LSWI joint with support veetor maehine（SVM） achieved the best aeeuraey, the overall elassification accuracy was 74. 18% and the Kappa coefficient was 0.60. The results showed that the support vector machine algorithm outperformed the maximum likelihood algorithm and the random forest algorithm. In Heilongjiang Provinee, the best period for sorting rice is the transplanting period in May, which can be characterized by LSWI. Theoptimal period for distinguishing between eorn and soybean was from the end of September to the beginning of October, which was the period when the soybean was harvested and the corn was not, and the optimal classification feature was EVI. This method provided a reference value for eropped area mapping in other agrieultural regions.
作者
黄健熙
侯矞焯
武洪峰
刘峻明
朱德海
HUANG Jianxi HOU Yuzhuo WU Hongfeng LIU Junming ZHU Dehai(College of Information and Electrical Engineering, China Agricultural University, Beijing 100083, China The Institute of Scientific and Technical Information, Heilongfiang Academy of Land Reclamation Sciences, Harbin 150036, China)
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2017年第10期142-147,285,共7页
Transactions of the Chinese Society for Agricultural Machinery
基金
国家自然科学基金项目(41671418
41471342
41371326)
国家高技术研究发展计划(863计划)项目(2013AA10230103)
