Search

Small ocean eddies are observed in fine-resolution satellite imagery within the coastal regions and ice margins of the world's oceans. The derivation of surface velocity fields of these features would enable geophysical estimates of the kinematic energy contained within the eddies, leading to improved understanding of the role these eddies play in ocean circulation. In this paper, we develop a method to derive surface velocity fields of small ocean eddies using fine resolution, multi-sensor imagery obtained over short time intervals. We consider the evolving image to be embedded in a deformable medium, and perform image matching using nonlinear viscous fluid registration model. We employ sum of squared differences and mutual information as matching functionals. The physical continuum equation is solved using an efficient multigrid full approximation scheme. We test our method using synthetic aperture radar (SAR) image pairs and show that it produces promising results.

Classic tie-point detection algorithms such as the Scale Invariant Feature Transform (SIFT) show their limitations when the images contain drastic changes or repetitive patterns. This is especially evident when considering multi-temporal series of images for change detection. In order to overcome this limitation we propose a new algorithm, the Affine Parameters Estimation by Random Sampling (APERS), which detects the outliers in a given set of matched points. This is accomplished by estimating the global affine transform defined by the largest subset of points and by detecting the points which are not coherent (outliers) with the transform. Comparisons with state-of-the-art methods such as GroupSAC or ORSA demonstrate the higher performance of the proposed method. In particular, when the proportion of outliers varies between 60% and 90% APERS is able to reject all the outliers while the others fail. Examples with real images and a shaded Digital Elevation Model are provided.

Applications such as change detection and digital elevation model extraction from optical images require a rigorous modeling of the acquisition geometry. We show that the unrecorded satellite jitter during image acquisition, and the uncertainties on the CCD arrays geometry are the current major limiting factors for applications requiring high accuracy. These artifacts are identified and quantified on several optical satellites, i.e., SPOT, ASTER, QuickBird, and HiRISE.

Recent methodological progress, Co-Registration of Optically Sensed Images and Correlation, outlined here, makes it possible to measure horizontal ground deformation from optical images on an operational basis, using the COSI-Corr software package. In particular, its sub-pixel capabilities allow for accurate mapping of surface ruptures and measurement of co-seismic offsets. We retrieved the fault rupture of the 2005 Mw 7.6 Kashmir earthquake from ASTER images, and we also present a dense mapping of the 1992 Mw 7.3 Landers earthquake of California, from the mosaicking of 30 pairs of aerial images.