摘要本文以浙江省长兴市作为研究区域,油菜作为主要研究对象,对油菜种植面积提取作相应研究。Landsat陆地卫星具有高时空分辨率,相较于其他卫星在农作物种植面积和长势监测应用较广泛,获取较简易。根据油菜物候规律和季相节律的差异性,选取相应时段的Landsat5 TM数据卫星影像, 借鉴分层信息提取法原理, 综合利用监督分类和非监督分类法,选择陆地卫星图像和分层信息提取的原理图,利用监督分类和非监督分类方法,具有人机交互目视解译和实地调查目标提取的信息长兴市油菜种植面积的,其中,监督分类的总体分类精度为98%,非监督分类的总体分类精度为89.684%。结果表明:Landsat5陆地卫星可用于提取油菜种植面积和生长发育监测,以满足农情监测的需求,并利用多时相遥感图像能有效增加的信息量,并信息互补有助于提高监测精度。61583
毕业论文关键词:油菜;Landsat卫星;TM数据;种植面积
In this paper, Changxing, Zhejiang is taken as the survey region and rape is taken as major study object to carry out related research on the rape planting area. The Landsat with high temporal-spatial resolution is more widely applied in supervision of crop planting area and growth and is easier to acquire compared to other satellites. In accordance with difference between the phenological rule and seasonal rhythm of rape,Landsat-5 satellite image of corresponding period is selected to use.Learning from the theory of hierarchical information extraction, supervised classification and unsupervised classification are combined to select the Landsat image and schematic diagram of hierarchical information extraction. Supervised classification and unsupervised classification are used. The visual interpretation with man-machine interaction and field survey target are adopted to extract the information of the rape planting area in Changxing. The overall classification accuracy of supervised classification is 98% while the overall classification accuracy of unsupervised classification is 89.684%. It turns out that Landsat5 can be used to extract the supervision of rape planting area and growth to meet the demand of crop condition monitoring. Multi-temporal remote sensing image can be used to effectively increase amount of information. In addition, the complementary information is helpful to improve the monitoring accuracy.
Keywords: Rape; Landsat satellite; TM data; acreage
目录
油菜种植面积提取 1
1引言 3
2 选题背景与研究意义 4
2.1 选题背景 4
2.2 研究区域 4
2.4 研究意义 5
3 数据与方法 5
3.1 数据源 5
3.2 ENVI软件介绍 5
3.3 研究方法 5
3.4 数据处理方法 6
3.4.1 手动辐射定标 6
3.4.2 大气校正 8
3.4.3 最大似然法 9
3.4.4 非监督分类 12
4 结果分析 15
致 谢 17
1引言
中国是世界上油菜生产种植的大国,油菜是我国五种油料作物中贡献最突出的一种作物, 是继玉米、水稻、大豆和小麦之后的第五大优势作物,它不仅是是主要的食用油源和蛋白质饲料,也是重要的工业原料,油菜的种植面积占中国油料作物种植总面积的40%以上,约占世界的30%以上,油菜的面积和产量都是世界首位。为了帮助国家有关部门及时有效的掌握作物的种植面积、分布和产量等信息,我们需要通过利用遥感卫星,通过卫星我们能够大面积、动态、准确、高效地获取作物空间分布和面积的优势。利用得到的信息能够帮助农业部门指导农业生产, 调整农业结构,为区域油菜的产量估计和预测提供可靠依据。利用遥感技术提取作物种植面积是农情监测的主要内容之一, 其重点在于作物种类的识别和面积量算的精度。因此, 可靠的数据源和准确、有效的油菜种植面积提取方法是油菜面积遥感估算的关键。 Landsat5陆地卫星ENVI油菜种植面积提取:http://www.751com.cn/guanli/lunwen_67400.html