一种面向对象结合变差函数的高分辨率遥感影像茶种植区自动提取方法

针对传统的茶区提取主要依靠人工野外勘测方法,需要耗费大量的人力物力,时效性差,精度低,不能及时有效地获取茶区空间分布信息,同时茶区在光谱特征上与林地、梯田等具有极强的相似性,茶种植区的遥感识别工作难度高等问题,提出了一种面向对象结合变差函数的茶种植区自动提取方法。为了解与动态监测茶的种植情况,选取位于贵州省铜仁市的4块矩形区域作为研究区。首先,采用面向对象的方法,构建分类规则集,剔除包括道路、建筑物、水体在内的非植被区域;然后,利用茶种植区与其他植被区域的变差函数纹理特征差异,构建决策树分类模型,选择最合适的纹理提取窗口尺寸,最终实现茶种植区域信息自动提取。实验结果表明,该方法可以准确识别茶种植区的空间分布情况(生产者精度为74.50%,用户精度为83.69%),为茶种植区自动提取提供了一种新的有效方法,可以为茶种植区监测与管理提供有效手段与科学依据。

An Object-oriented and Variogram Based Method for Automatic Extraction of Tea Planting Area from High Resolution Remote Sensing Imagery

The traditional tea area extraction mainly depends on the artificial field survey method,which needs a lot of manpower and material resources,poor timeliness and low precision,and cannot obtain the spatial distribution information of tea area timely and effectively.At the same time,the spectral characteristics of tea area are very similar to forest land and terrace,and the remote sensing identification of tea planting area is difficult.In view of this,an object-oriented and variogram based method for automatic extraction of tea growing areas is proposed in this paper.In order to understand and dynamically monitor the tea planting situation,four rectangular areas located in Tongren city of Guizhou province are selected as the research area.Firstly,an object-oriented method is used to construct a classification rule set,which excludes non-vegetated areas including roads,buildings,and water bodies.And then,use the difference in texture features of the tea plantation area and other vegetation areas to build a decision tree classification model.Finally,select the most appropriate texture extraction window size and realize automatic extraction of tea planting area information.Experimental results show that the proposed method can accurately identify the spatial distribution of tea planting areas,with a producer accuracy of 74.50%and a user accuracy of 83.69%.It provides a new and effective method for automatic extraction of tea growing areas,which can provide effective means and scientific basis for monitoring and management of tea growing area.