As the sensor technique develops rapidly, especially the high-precision commercial satellite, Rational Function Model (RFM) has received increasingly more acceptance because of its simple form, convenience for using, low requirement for specialized knowledge and not depending on the imaging parameters of specific satellite. Nevertheless, the quality of applying the model depends on the precision of parameters. Because of the ill-conditioned equation during solving the RFM, traditional Least square (LS) method can not always offer high-precision solution, therefore, it is essential to research better methods to provide theory basis for wider application of RFM. In the paper, truncation singular value decomposition, Genetic Algorithm and various biased estimation methods are applied to solve the parameters of RFM, and detailed analysis is made about the results of different algorithms that they all can solve the ill-conditioned equation well and improve the precision. The conclusion is that truncation singular value decomposition, Genetic Algorithm and using biased estimation method reasonably can lead to high-precision parameters but Genetic Algorithm remains to be improved due to its extensive time.

Date Published: 14 November 2007
PDF: 7 pages
Proc. SPIE 6790, MIPPR 2007: Remote Sensing and GIS Data Processing and Applications; and Innovative Multispectral Technology and Applications, 67901R (14 November 2007); doi: 10.1117/12.749318

Published in SPIE Proceedings Vol. 6790:
MIPPR 2007: Remote Sensing and GIS Data Processing and Applications; and Innovative Multispectral Technology and Applications
Yongji Wang; Jun Li; Bangjun Lei; Chao Wang; Liang-Pei Zhang; Jing-Yu Yang, Editor(s)