Advanced Dinsar Based on Coherent Pixels: Development and Results Using CPT Technique

Jesús Domínguez(1) , Rosana Romero(1) , Daniel Carrasco(1) , Antonio Martínez(1) , Jordi J. Mallorquí(2) , Pablo Blanco(2) , David Navarrete(2) , and Sergi Duque(2)

(1) Indra Espacio, Mar Egeo 4, 28830,San Fernando de Henares (Madrid), Spain
(2) Universitat Politècnica de Catalunya (UPC), D3-Campus Nord-UPC, 08034, Barcelona, Spain


This paper presents the activities undertaken and the results obtained in the project “Development of algorithms for the exploitation of ERS – ENVISAT using the Coherent Pixels Technique”. The project objective was to develop an advanced interferometric technique based on the coherence stability and study the applicability in a group of joint interferograms obtained from both ERS and ENVISAT images. The work has been carried on by Indra and UPC in the frame of an ESA contract.

The Coherent Pixels Technique (CPT) algorithm is capable of extracting the linear and non-linear movement of a given set of interferograms, as well as the DEM error and atmospheric artefacts. The pixel selection criteria is based in the coherence stability. As a consequence, the final product will have lower resolution than the original images. The characteristics of the algorithm make possible to obtain good results with a small number of images, compared with other advanced DINSAR techniques. The advantages and disadvantages of the algorithm will be discussed in the paper.

A Graphical User Interface has been developed in order to provide a friendly handling of data, involving an automation of the INSAR processing chain (batching), the implementation and enhancements of the CPT algorithm, a browser tool to allow the estimation of the number and density of the coherent pixels, and a final results viewer. The GUI will be presented, showing some practical details.

Two main scenarios have been used for the validation of the algorithm in the project: Paris and Madrid. They have been selected by their different characteristics and type of deformation phenomena present. The city of Paris is a large urban area where well reported very non-linear deformation patterns are present. The area of Madrid is above a big aquifer where recent variations in the water level have occurred, and represented for us the possibility of using simultaneously a small set of ascending and descending images. The deformation results obtained during validation tests in the different scenarios will be presented, proving that CPT algorithm is a good tool for extracting terrain deformation with a reduced number of images.


Workshop presentation

Full paper


  Higher level                 Last modified: 07.10.03