Title :
Cause analysis and accuracy improvement of confusing land cover types in China geography census
Author :
Zhihong Gao ; Xu Zhou
Author_Institution :
Nat. Geomatics Center of China, Beijing, China
Abstract :
Geographic national conditions monitoring is an important new mission of China. China geography census is a major national survey of national condition and power, and it is a basic work to access the fundamental natural, ecological and human activities information of land surface. Land cover classification data is an important part of the China geography census, and it is the base of the subsequent statistical analysis. So the accuracy of the classification result will directly affect the data quality. In order to solve this problem, an extensive collection of typical land cover examples and problems is carried out, and the 94 confusing land cover groups are given. Meanwhile, the confusing causes are analyzed comprehensively, the association analysis maps are provide. And 6 suggestions provided in order to improve the final classification accuracy. The results of this study can provide scientific reference for the land cover classification of China geography census.
Keywords :
geography; geophysical image processing; image classification; land cover; remote sensing; statistical analysis; China geography census; association analysis maps; cause analysis; confusing land cover types; data quality; ecological activity information; geographic national conditions monitoring; human activity information; land cover classification data; land surface; natural activity information; statistical analysis; Accuracy; Buildings; Condition monitoring; Geography; Land surface; Roads; Statistical analysis; China geography census; confusing; data quality; land cover; remote sensing imagery;
Conference_Titel :
Agro-geoinformatics (Agro-geoinformatics 2014), Third International Conference on
Conference_Location :
Beijing
DOI :
10.1109/Agro-Geoinformatics.2014.6910660