Estimacion y corrección de errores orbitales en interferometría SAR
Correu electrònic de l'autorjuan.carlos.mateo.sanchez93gmail.com
Realitzat a/ambDares Technology
Tipus de documentTreball Final de Grau
Condicions d'accésAccés restringit per acord de confidencialitat
The present project is based on InSAR (Interferometric Synthetic Aperture Radar) technique. This technique allows the detection of anomalies on the Earth's surface by monitoring extended areas with SAR images. InSAR requires high performance software tools in order to obtain optimum resolution information (in particular the detection of terrain deformation at mm scale). One of the critical problems of InSAR techniques is the presence of errors in the generated images. These errors come from different sources and may significantly decrease the reliability of results. In the present project we will focus on errors caused by inaccuracies in the transmission of the satellite's orbital information, in particular, due to actual displacements during the acquisition of images which should have been taken from the same position and at the same velocity. We describe the bases of SAR and then focus on the details of the techniques which are actually used to mitigate the above mentioned errors. Using different simulations we argue that these techniques are insufficient, as they do not provide error correction at a satisfactory level. We then propose the necessity of an alternative more efficient method of error correction and describe the algorithm developed to improve performance in this respect. We present details of its evolution and the efficiency attained by showing results of corrections applied to real cases. We compare results from the original and the new method proposed in order to verify the actual improvement achieved. Finally we draw the main conclusions from our work in terms of goal accomplishment, and outline the future work to develop if our proposed error correction technique is to be used as part of the quality control process in the release of InSAR images.