›› 2005, Vol. 7 ›› Issue (3): 97-103.

• ARTICLES • Previous Articles     Next Articles

Research on Multi-source Remote Sensing Information Fusion Application

YUAN Jinguo, WANG Wei   

  1. College of Resource and Environmental Sciences, Hebei Normal University, Shijiazhuang 050016, China
  • Received:2004-02-19 Revised:2004-02-29 Online:2005-09-25 Published:2005-09-25

Abstract: Multi-source remote sensing data fusion is the development trend of remote sensing technology in depth. This paper analyzes in detail algorithmic application characteristics of multi-source remote sensing data from three levels of pixel-based, feature-based and decision-based fusion processings. Take Fengning County for example, specific applications of remote sensing data fusion methods in information extraction are illuminated. The data used in this study is firstly pre-processed, then the principal components of Landsat TM data in 1999 are analyzed, the first three principal components account for 97.8% of the total information, the resulted image of inversed principal components transformation is clearer and has more abundant levels. To extract information from remote sensing image, we select the fusion image from Landsat TM pan and multi-spectral bands after principal components transformation, color composition scheme of bands 4, 3, 2 and bands 5, 4, 3, and vegetation index and greenness index after tasseled cap transformation are analyzed, the remote sensing image information fusion with DEM and spatial data of GIS database can also improve the accuracy of remote sensing information extraction. Problems to be resolved and future direction of multi-source remote sensing data fusion are put forward.

Key words: multi-source data, remote sensing, information fusion