基于GF-1影像面向对象分类方法的水稻种植信息提取研究

发布时间：2019-01-23 17:01

【摘要】：应用遥感技术提取水稻种植信息是农业遥感的重要内容。GF-1卫星WFV数据为农业信息提取提供了新的途径,面向对象的分类方法是遥感解译的重要方法。本研究以扬州市为研究区域,基于GF-1影像WFV数据,采用面向对象的分类方法,提取水稻种植信息,并实地调查验证试验结果,试图探讨GF-1数据面向对象分类方法在水稻种植信息提取中的可行性与影响提取精度的因素。结果表明,应用GF-1数据,采用面向对象的分类方法能够很好地完成扬州市水稻种植信息的提取,2016年扬州市有水稻种植面积214 524 hm~2,总体精度达到98.5%,Kappa系数0.95,面积精度达97.5%;实地考察能够提高提取精度,地形破碎程度越低,提取精度越高。
[Abstract]:Using remote sensing technology to extract rice planting information is an important content of agricultural remote sensing. GF-1 satellite WFV data provide a new way for agricultural information extraction. Object-oriented classification method is an important method of remote sensing interpretation. In this study, Yangzhou City was used as the research area. Based on the WFV data of GF-1 image, the information of rice planting was extracted by using the object-oriented classification method, and the results of the experiment were verified by field investigation. This paper attempts to explore the feasibility of object oriented classification of GF-1 data in rice planting information extraction and the factors affecting the extraction accuracy. The results show that the extraction of rice planting information in Yangzhou can be accomplished well by using GF-1 data and object-oriented classification method. The total precision of 214,524 hm~2, of rice planting area in Yangzhou in 2016 is 98.5%. The Kappa coefficient is 0.95and the area precision is 97.5. Field investigation can improve the extraction accuracy, the lower the degree of topographic fragmentation, the higher the extraction accuracy.
【作者单位】：扬州市耕地质量保护站;福建师范大学地理科学学院;中国科学院地理科学与资源研究所/资源与环境信息系统国家重点实验室;扬州地恒科技有限公司;
【基金】：国家重点研发计划项目(2016YFD0200301) 国家重大科技专项项目“新能源评估研究示范”(30-Y30B13-9003-14/16-04) 农业部耕地质量保护项目(农财发[2016]35)
【分类号】：S127;S511

【相似文献】