面向农用土地现状调查的多源数据遥感解译:一种基于WebGIS的方法

发布时间：2018-05-04 02:11

本文选题：土地利用调查 + 农业遥感　；参考：《华东师范大学》2015年硕士论文

【摘要】：目视解译是大比例尺遥感土地利用分类的基本方法,长期以来存在着效率和精度方面的问题。一个重要原因是解译过程受限于传统纸质操作平台,无法有效整合多源异质数据来支持复合逻辑判断。WebGIS的使用可以弥补这一方面的缺陷,即通过对在线/线异源异质相关数据的空间整合,以及多时相遥感图像的组合,搜集更为充分的视觉证据,最大程度地降低地类判读的不确定性,并能即时将矢量化判读结果直接整合进入数据库。本研究以农业用地现状的大比例尺遥感调查为切入点,研究基于WebGIS平台的多源异质数据的空间整合方法与技术,探索和建立精细地类目视判别的多元复合证据逻辑推理规则,旨在提高可分类别数目和分类精度。本文的科学问题是,对于在光谱和形态上无法识别的地物,如何通过多证据复合推理来建立其类别语义的可信度。这里牵涉到对研究区域的农业文化和实践特点的知识运用,并非单纯的视觉判断。研究结果将提炼出用于遥感判读专家系统／知识库的判读规则。本文首先阐述土地利用调查和遥感影像目视解译相关基础理论,并简述3S技术(即地理信息系统、遥感及全球定位系统)在土地利用调查中的应用。其次,从实际应用需求出发,参与设计了农业用地现状数据在线采集系统。以遥感影像目视解译理论为基础,结合该系统的特点,通过整合多源辅助信息,研究和提出了一种基于WebGIS的农业用地现状数据调查方法,这是本研究的方法论基础。最后,以上海市为例,对上述解译方法进行了实例探究。在课题组完成上海市农业用地现状数据在线采集工作的基础上,进行了精度验证,结果数据质量很高,证明了本研究方法的合理性。另外,作者通过参与室内判读和外业调研工作,进一步补充和完善了基于航拍图像的上海市农业用地判读规则,为后期计算机自动解译算法设计提供参考。
[Abstract]:Visual interpretation is the basic method of land-use classification for large scale remote sensing. There have been problems in efficiency and accuracy for a long time. One of the important reasons is that the process of interpretation is limited to the traditional paper operating platform. It can not effectively integrate multi source heterogeneous data to support the use of complex logic judgment.WebGIS, which can make up for this problem. In this research, the large scale remote sensing of the current situation of agricultural land can be obtained by integrating the spatial integration of the heterogeneous and heterogeneous data of the online / line and the combination of the multi temporal remote sensing images to collect more sufficient visual evidence to minimize the uncertainty of the class interpretation and directly integrate the vectorization interpretation results into the database. As a breakthrough point, the spatial integration method and technology of multi source heterogeneous data based on WebGIS platform are studied, and the logical reasoning rules of multiple compound evidence for fine visual discrimination are explored and established to improve the number of classifications and classification accuracy. The scientific problem in this paper is for objects that can not be identified in spectral and form, such as There is no simple visual judgement on the application of knowledge to the agricultural culture and practice characteristics of the study area. The results will extract the interpretation rules for the remote sensing interpretation expert system / knowledge base. This paper first expounds the land use survey and remote sensing. The application of 3S Technology (geographic information system, remote sensing and global positioning system) in land use survey is described. Secondly, the online acquisition system of agricultural land use status data is designed from the actual application requirements. Based on the theory of remote sensing visual interpretation, the characteristics of the system are combined. Through the integration of multi source auxiliary information, this paper studies and puts forward a survey method of agricultural land use status data based on WebGIS, which is the basis of the methodology of this study. Finally, taking Shanghai as an example, a case study is carried out on the above interpretation method. On the basis of the task group completing the online data collection of the current situation of agricultural land in Shanghai, the project is carried out. The quality of the result data is very high, which proves the rationality of this method. In addition, the author further supplemented and perfected Shanghai agricultural land interpretation rules based on aerial images, and provided reference for the design of automatic interpretation algorithm in the later period.

【学位授予单位】：华东师范大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P208;P237

【相似文献】