| 基于几何语义知识的冬小麦自动分类 |
| |
| 引用本文： | 王利民,刘佳,邵杰,杨福刚,季富华,姚保民.基于几何语义知识的冬小麦自动分类[J].中国农学通报,2019,35(19):120-130. |
| |
| 作者姓名： | 王利民 刘佳 邵杰 杨福刚 季富华 姚保民 |
| |
| 作者单位： | 中国农业科学院农业资源与农业区划研究所 |
| |
| 基金项目： | 高分辨率对地观测系统重大专项（民用部分）(09-Y20A05-9001-17/18)。 |
| |
| 摘 要： | 【研究目的】为了实现遥感影像的作物自动分类，并探索空间信息在分类中作用，【方法】本文提出结合光谱和空间信息的作物分类方法。首先，借助光谱信息实现地物初始分割，然后以目标作物历史空间分布为语义约束，根据隶属度提取目标作物。最后，在多时相遥感影像条件下，以冬小麦为目标作物进行了方法的验证，【结果】结果显示，本文方法可实现冬小麦自动提取与识别，总体精度为95.33%，Kappa系数为0.90，可满足农情监测的实际需求。另外，在单时相遥感影像条件下，本文结合几何语义知识的作物分类精度也达到了较高水平。【结论】相对于遥感影像单一光谱信息的分类方法，本文方法利用了作物空间信息，不仅能满足精度要求，还实现了分类的自动化，对工程化应用具有一定的参考价值。
|
| 关 键 词： | 薏苡 薏苡 高原地区 品种（系） 种植密度 追肥量 高产栽培 |
| 收稿时间： | 2019/1/7 0:00:00 |
| 修稿时间： | 2019/6/12 0:00:00 |
Automatic Classification of Winter Wheat: Based on Geometry Semantic Knowledge |
| |
| Abstract: | Objective]The purpose of the study is to realize crop automatic classification based on remote sensing images, and to explore the roles of spatial information in crop classification, Method]The paper proposes a crop classification method with the combination of spectral and spatial information. Firstly, the spectral information is utilized to realize the initial division of ground objects; secondly, by taking the historical spatial distribution of target crops as semantic constraints, the target crops are extracted based on degree of membership; Finally, under the condition of multi temporal remote sensing images, the method is verified by taking winter wheat as target crop. Result]The result shows that, the method proposed in this paper can realize winter wheat automatic extraction and identification, with the overall accuracy of 95.33%，and Kappa coefficient of 0.90. The can meet the practical demand of agricultural condition monitoring. In addition, under the condition of mono temporal remote sensing image, the crop classification accuracy of the proposed method with the combination of geometric semantic knowledge also has reached a high level. Conclusion]Compared with the remote sensing image classification method with single spectral information, the method proposed in this paper makes use of crop spatial information, and it not only meet requirement of accuracy, but also realizes automatic classification. It has certain reference value for engineering application. |
| |
| Keywords: | Geometricy semantic knowledge ISODATA Winter wheat Automatic classification |
| 本文献已被 CNKI 等数据库收录！ |
| 点击此处可从《中国农学通报》浏览原始摘要信息 |
|
点击此处可从《中国农学通报》下载全文 |
|
