The Automatic Extraction of DEM Using Modified RFM of IKONOS Stereo Imagery

EasyChair Preprint 2830

Abstract

The launch of extremely high-resolution sensor satellites has paved the way for many researchers to explore the potential gain in producing Digital Elevation Model (DEM) from stereo imagery. This paper deals with the issue of extracting DEM automatically from IKONOS stereo-pair of imagery using modified the Rational Function Model (RFM). Since high resolution satellite imagery companies are not provide precise sensor information and orbit parameters for various reasons. Thus, the use of RFM as a generic sensor model has become an alternative to accurate sensor models. This model uses the Rational Polynomial Coefficients (RPCs) provided with the images describing the relationship between ground space and image space. However, the RPCs supplied by the vendors may not always represent well the true imaging process and this affects the accuracy of DEM. Therefore, RPCs modification is better choice for obtaining DEM with reasonable accuracy. The study was conducted on a stereo pair of IKONOS imagery over an area in North of Khartoum (Sudan), covering about 30 sq-km. A well distributed set of 12 ground points was selected over stereo images and then surveyed using Differential Global Positioning System (DGPS) static technique, these points are divided into control and check points to assess the accuracy. DEM generation automatically depends on a series of processes within the imagine photogrammetry software package that were used in this research with the modified RFM. Results of the experiment showed that the accuracy of extracted DEM by modified RFM has a very good consistency and within a 2 pixel-level using a few Ground Control Points (GCPs). Finally, the results of this paper are mainly motivating for professionals from different disciplines who need to use a DEM for planning and transportation issues as well as using high-resolution stereo images to replace the traditional aerial photography in generating DEM, which limits the cost and time consuming.

Keyphrases: DEM, GCPs, IKONOS, Imagine Photogrammetry Software, RFM, RPCs

@booklet{EasyChair:2830,
  author    = {Tamer Saleh},
  title     = {The Automatic Extraction of DEM Using Modified RFM of IKONOS Stereo Imagery},
  howpublished = {EasyChair Preprint 2830},
  year      = {EasyChair, 2020}}