TY - JOUR

T1 - Automatic Generation of Orthorectified High Resolution Satellite Imagery

T2 - a Case Study for Saudi Arabia

AU - Heipke, Christian

AU - Alrajhi, Muhamad

PY - 2016/2

Y1 - 2016/2

N2 - The Ministry of Municipal and Rural Affairs (MOMRA) is responsible for the production, maintenance and delivery of accurate geospatial data for all urban and rural settlements in the Kingdom of Saudi Arabia. The fast urbanization requires up-to-date information, which is a major challenge. High resolution satellite imagery presents a novel and promising data resource for geospatial data update. It is well-known that a satellite image, if georeferenced using the vendor-provided rational polynomial coefficients (RPC), often displays X- and Y-coordinate biases of several pixels. Ground control points (GCP) are typically required to achieve an accuracy at pixel level. The collection of the 3D coordinates of GCP and their corresponding image coordinates is a time consuming and cost intensive manual process. The primary objective of this research is to use image matching between existing orthophotos and the satellite images, georeferenced with RPC instead of GCP, in order to automate and speed up the orthorectification process. Based on a series of practical experiments using imagery from GeoEye and IKONOS, the potential of automated matching between aerial and satellite images using the well known Speeded-Up Robust Features (SURF) algorithm for the described task is investigated. The matched points serve as a basis for transforming the satellite orthophoto to the aerial orthophoto using a 2D affine transformation. This research has led to the development of a simple and efficient tool for satellite imagery with a ground sampling distance of 50 cm to 1 m to be used for updating geospatial information that meets MOMRA accuracy standards for 1:10,000 scale mapping. The process completely eliminates the need for GCP. About 12 to 15 satellite images are routinely being processed on a workstation in a single day. The implementation of this tool at MOM-RA greatly enhances its ability to quickly respond to urgent needs for updated geospatial data.

AB - The Ministry of Municipal and Rural Affairs (MOMRA) is responsible for the production, maintenance and delivery of accurate geospatial data for all urban and rural settlements in the Kingdom of Saudi Arabia. The fast urbanization requires up-to-date information, which is a major challenge. High resolution satellite imagery presents a novel and promising data resource for geospatial data update. It is well-known that a satellite image, if georeferenced using the vendor-provided rational polynomial coefficients (RPC), often displays X- and Y-coordinate biases of several pixels. Ground control points (GCP) are typically required to achieve an accuracy at pixel level. The collection of the 3D coordinates of GCP and their corresponding image coordinates is a time consuming and cost intensive manual process. The primary objective of this research is to use image matching between existing orthophotos and the satellite images, georeferenced with RPC instead of GCP, in order to automate and speed up the orthorectification process. Based on a series of practical experiments using imagery from GeoEye and IKONOS, the potential of automated matching between aerial and satellite images using the well known Speeded-Up Robust Features (SURF) algorithm for the described task is investigated. The matched points serve as a basis for transforming the satellite orthophoto to the aerial orthophoto using a 2D affine transformation. This research has led to the development of a simple and efficient tool for satellite imagery with a ground sampling distance of 50 cm to 1 m to be used for updating geospatial information that meets MOMRA accuracy standards for 1:10,000 scale mapping. The process completely eliminates the need for GCP. About 12 to 15 satellite images are routinely being processed on a workstation in a single day. The implementation of this tool at MOM-RA greatly enhances its ability to quickly respond to urgent needs for updated geospatial data.

KW - High resolution

KW - Image matching

KW - Orthophotos

KW - Satellite imagery

UR - http://www.scopus.com/inward/record.url?scp=84971671375&partnerID=8YFLogxK

U2 - 10.1127/pfg/2016/0285

DO - 10.1127/pfg/2016/0285

M3 - Article

AN - SCOPUS:84971671375

VL - 2016

SP - 5

EP - 15

JO - Photogrammetrie, Fernerkundung, Geoinformation

JF - Photogrammetrie, Fernerkundung, Geoinformation

SN - 1432-8364

IS - 1

ER -