Empirical Mode Decomposition of Data Analysis for DTM

• Main Authors: Horng-Shin Jiang, 江鴻鑫
• Other Authors: Zhen-Wwi Li
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/00359990698405658083

碩士 === 國防大學中正理工學院 === 軍事工程研究所 === 94 === Digital Terrain Model（DTM）is a powerful tool to describe topography trully, it is not only applied to generate contour lines, orthophotograph, virtual reality, 3-D city simulation…etc., but also provides the platform of image database and computer wargame exercise to build synthetic battlefield reguired forCommand, Control, Communication, Computer, Intelligence, Surveillance, Reconnaissance（C4ISR）in military. Besides, Many fields in geomatics depend largely on DTM, such as country planning, mining industry, military simulation, geomorphology, ecology, hydrology…etc. The Empirical Mode Decomposition method（EMD）has been proved to be able to improve the data analysis of non-period, nonlinear and non-stationary signals. EMD has the characteristics of intuitive, direct, a posteriori and adaptive while instantaneous frequency changes can be observed. In this paper Normalized Cross-Correlation（NCC）has been used to give better Stop Condition to make sure convergence of Intrinsic Mode Function（IMF）. Meanwhile, Mirror Extend is used to improve End Effect to reserve data reality. Finally, EMD is applied resolve DTM signals together with better combination of different interpolation methods to increase potency.