Geometric Correction and Automatic Dem Extraction of Worldview Stereo Data

The successful operation of the WorldView-1and 2 satellites have enabled advanced high-accuracy mapping using high-resolution satellite images, with and without the use of ground control points (GCPs). The stereo capability of the WorldView sensor also provides the opportunity to extract high-resolution digital elevation models (DEMs). In this paper we will examine the following areas: (1) Each WorldView data is distributed with Rational Polynomial Coefficients (RPCs). The coefficients can be used to correct the WorldView data using Rational Function Method (RFM). What is the minimum number of GCPs required to achieve the best geometric accuracy? (2) WorldView data is distributed in Basic 1B and Ortho-Ready Standard (OR2A) formats. Which format should be used to produce the best accuracy in geometric correction? (3) In addition to the high-resolution panchromatic sensor, WorldView-2 has eight multispectral bands which can be used for many applications such as classification and pansharpening. We will examine the pansharpening of WorldView-2 data using an algorithm developed at the University of New Brunswick, Canada. (4) Different stereo pairs can be acquired by the WorldView satellite. What is the best stereo viewing geometry for automatic DEM extraction? This paper tested different WorldView satellite data to provide the answers to the above questions.