Translator Disclaimer
Paper
30 October 2009 Mapping forest cover using aircraft imagery in Qisha peninsula, South China
Author Affiliations +
Proceedings Volume 7498, MIPPR 2009: Remote Sensing and GIS Data Processing and Other Applications; 749817 (2009) https://doi.org/10.1117/12.832451
Event: Sixth International Symposium on Multispectral Image Processing and Pattern Recognition, 2009, Yichang, China
Abstract
Land use and forest cover products in highly dynamic tropical ecosystems lack the detail needed for natural resource management and monitoring at the regional level. The aerial photography provides effective methods to combine local forest cover data for land cover mapping and land resource management. In this paper we use aircraft imagery at the regional level over a characteristic location of tropical woodlands in Guangxi, China. The performances of maximum likelihood classification and Decision tree supervised classification were assessed. The most consistent results were achieved using decision tree analysis of aircraft images. This method provided better accurate classification for mangroves, broad-leaved forest, coniferous forest and shrub and the average classification accuracy is above 86%. The aircraft image provided a more accurate classification for the dense forest cover class. The selection of the right image dates proved to be critical for different forest type recognition. It make them promising options for rapid and inexpensive forest cover mapping in regions of high environmental variability such as tropical coastal zone.
© (2009) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Rong Yu, Bofeng Cai, Shanjing Zhong, and Hui Zen "Mapping forest cover using aircraft imagery in Qisha peninsula, South China", Proc. SPIE 7498, MIPPR 2009: Remote Sensing and GIS Data Processing and Other Applications, 749817 (30 October 2009); https://doi.org/10.1117/12.832451
PROCEEDINGS
5 PAGES


SHARE
Advertisement
Advertisement
Back to Top