Science Service System
You are here : Home : Proposals_Summary

Summary of Proposal LAN1085

TitleAssessment of fusion of optical and radar high resolution satellite images for generating large scale map
Investigator goodarz, farhad - MSc student of photogrammetry and remote sensing, K.N.toosi.university
Team Member
dr dehghani, Maryam - group of Remote Sensing Engineering, K.N.Toosi University of Technology, Faculty of Geodesy & Geomatics Engineering,Geological Survey of Iran (GSI)
dr varshosaz, Masood - group of photogrammetry and remote sensing, Faculty of Geodesy and Geomatics Eng, K. N. Toosi University of technology
SummaryHigh-resolution satellite imagery (HRSI) is being increasingly employed for large-scale topographic mapping, and especially for geodatabase up-dating. Geoeye-1 has exhibited georeferencing accuracy as high as 0.1m (0.2 pixel) in planimetry and 0.25m (0.5 pixel) inheight in very well controlled metric evaluation tests involving stereo image pairs and triplets.Generation of digital elevation models (DEM) is a main issue of satellite remote sensing and is needed for many applications of satellite imagery. Today, both satellite stereoscopy and interferometry, exist in parallel and are applied independent from each other . Objectives: Topographic maps including 3 dimensional (3D) information can be generated using different methods. One of the most widespread approaches is to use the optical stereo remotely sensed images such as IKONOS and Geoeye-1 data to extract 3D coordinates of the objects on the earth surface. However, there are sometimes problems in the optical stereo images decreasing the geometry and radiometric quality of the image such as cloud covers, radiometric differences and dependency on daylight. This fact leads to matching problems and reduces the accuracy of the products. On the other hand, other techniques such as SAR interferometry (InSAR) can mitigate the above-mentioned problems while being influenced by inherent sources of errors such as layover, foreshortening and speckle noise. The main purposes of this study as a MSc thesis are as follows: (i)To assess the accuracy of Digital Elevation Models (DEM) extracted from both radar images (InSAR) as well as optical stereo images: it will be tried to generate DEM from high-resolution TerraSAR-X images using InSAR. Meanwhile high resolution optical Geoeye-1 stereo images are employed in order to extract DEM from the same study area. The accuracy of both DEMs is evaluated using a 1:2000 reference topographic map. (ii)To combine the 2D planar information from a mono optic image with the height information extracted from TerraSAR-X in order to generate topographic map in those cases that we are not able to produce DEM from optical data. One of the aspects of merging information would be to transform the radar imaging acquisition geometry into that of the optic image. Alongside we would be able to employ radar amplitude image as a complementary source to extract the 2D planar information. Method: Conventional SAR Interferometry (InSAR) will be used in order to generate DEM from TerraSAR-x data. The unwrapped coherent interferograms include the topographic information. In DEM generation using the optical images, the orbital parameter model and rational function can be applied into the Level 1A and Level 1B, respectively. Data requirements: The required data for this study is: (1) TerraSAR-X high resolution spotlight data (2) Large-scale topographic map (1:2000,1:5000) (3) Geoeye-1 stereo images Deliverables: The outputs of the study are as follows: (1)The topographic products (including DEM, orthophoto large-scale topographic map generated from DEM and a mono optical image) (2)The evaluation results which will compare the potential of radar data, optical data and their combination in order to generate the large-scale topographic map. The project output will be delivered through a complete report to DLR. This project is financially supported by K. N. Toosi University of Technology, Faculty of Geodesy and Geomatics Engineering.

Back to list of proposals

DLR 2004-2016