基于J-M距离的多时相Sentinel-1农作物分类  

Time-series Classification of Sentinel-1 Data Based on J-M Distance

作　　者：韩冰冰 陈圣波[1] 曾庆鸿 孙士超 HAN Bing-bing;CHEN Sheng-bo;ZENG Qing-hong;SUN Shi-chao(College of Geo-Exploration Science and Technology,Jilin University,Changchun 130026,China;Chinese People s Liberation Army 93116 troops,Shenyang 110141,China)

机构地区：[1]吉林大学地球探测科学与技术学院,长春130026 [2]中国人民解放军93116部队,沈阳110141

出　　处：《科学技术与工程》2020年第17期6977-6982,共6页Science Technology and Engineering

基　　金：国家发改委东北地区培育和发展新兴产业三年行动计划中央预算内投资计划(吉发改投资[2016]512号);国防科工局、财政部关于高分专项专项省(自治区、市)域产业化应用项目(71-Y40G04-9001-15/18)。

摘　　要：为准确获取云雨天气较多地区的农作物分布信息,以敦化市西北部的黑市乡和额穆镇为研究区,以高分一号(GF-1)卫星宽视场(wide field view,WFV)传感器和哨兵一号(Sentinel-1)为数据源,通过WFV数据提取耕地范围,利用J-M(Jeffries-Matusita)距离可分性进行Sentinel-1影像最优分类时相组合的判断,根据选取Sentinel-1数据通过最近邻分类器多次循环对研究区主要农作物的分布信息进行提取。研究结果表明,进行多次循环产生的分类结果优于单次循环结果,循环次数达到5次结果趋于稳定,总体分类精度为84.23%。可见利用Sentinel-1数据进行敦化市农作物分类是可行的,多次循环的最近邻算法有利于获取更精准的农作物分布信息。In order to accurately obtain the distribution information of crops in areas with cloudy and rainy,Heishi Township and Emu Town in the northwest of Dunhua were selected as research sites,and the GF-1 wide field view(WFV)sensor and Sentinel-1 were utilized as the data source.The range of cultivated land was extracted through WFV data and the J-M distance feasibility was used to judge the optimal classification of the time-phase combination of Sentinel-1 image.According to the selection of Sentinel-1 data,the distribution information of main crops in the studied sites were extracted by the nearest neighbor classifier multiple cycles.The results show that the classification results produced by multiple cycles are better than the single cycle produced results.When the number of cycles reaches to 5 times,the results tend to stable,and the overall classification accuracy reaches to 84.23%.It can be seen that it is feasible to use Sentinel-1 data to classify crops in Dunhua.The multi-cycle nearest neighbor algorithm is beneficial to obtain more accurate crop distribution information.

关 键 词：农作物分类 Sentinel-1 J-M距离 最近邻分类器 

分 类 号：TP753[自动化与计算机技术—检测技术与自动化装置]