Abstract
This study presents a method to measure the displacement fields on the surface of planar objects with sub-pixel resolution, by combining image correlation with a differential technique. First, a coarse approximation of the pixel level displacement is obtained by cross-correlation (CC). Two consecutive images, taken before and after the application of a given deformation, are recursively split in sub-images, and the CC coefficient is used as the similarity measure. Secondly, a fine approximation is performed to assess the sub-pixel displacements by means of an optical flow method based on a differential technique. To validate the effectiveness and robustness of the proposed method, several numerical tests were carried out on computer-generated images. Moreover, real images from a static test were also processed for estimating the displacement resolution. The results were compared with those obtained by a commercial digital image correlation code. Both methods showed similar and reliable results according to the proposed tests.

Abstract
In this paper, a digital image correlation (DIC) method coupling cross-correlation with spatio-temporal differential techniques was proposed for assessing discontinuous displacement fields. The accuracy and robustness of the algorithm was assessed on a set of numerical tests by processing computer generated speckled-pattern images. Fracture mechanical tests in mode I were considered, in which both in-plane and out-of-plane rigid-body movements were taken into account. The ability for recovering the analytical asymptotic displacement field in mode I was analysed, and stress intensity factor, crack opening displacement and crack tip location were used as quantitative parameters for validation purposes. Throughout these tests, the results obtained with the proposed method were systematically compared to the ones from Aramis DIC-2D commercial code. Globally, the results computed from both methods are in good agreement with reference values. However, due to the high spatial resolution (point-wise characteristic), a better matching of the displacements in the neighbour of discontinuities could be obtained by the proposed method.

Abstract
In this paper, two Persistent Scatterer Interferometry (PSI) methodologies are compared in order to further understand their potential in the detection of surface deformation. A comparison of these two algorithms is a comparison of the two classes of PSI techniques available: coherence estimation based on a temporal model of deformation (represented by DePSI) and coherence estimation based on spatial correlation (represented by StaMPS). Despite the similarity between the results obtained from the application of the two independent PSI methodologies, significant differences in PS density and distribution were detected, motivating a comparative study between both techniques. We analyze which approach might be more appropriate for studying specific areas/environments, which is helpful in evaluating the benefits that could be derived from an integration of the two methodologies. Several experiments are performed to assess the sensitivity of both PSI approaches to different parameter settings and circumstances. The most significant differences in the processing chain of both procedures are then investigated. We apply both methodologies to the Granada Basin area (southern Spain) and realize that coherence does not improve significantly as function of the methodology applied. If oversampling is implemented in the StaMPS processing chain, the PS density increases so that the density in the urbanized areas is similar to the results provided by DePSI but in all the remaining covers the density is significantly higher. The general results provided by both approaches are very similar in the relative deformations estimated.

Abstract
The current paper proposes a new parametric local color correction technique. Initially, several color transfer functions are computed from the output of the mean shift color segmentation algorithm. Secondly, color influence maps are calculated. Finally, the contribution of every color transfer function is merged using the weights from the color influence maps. The proposed approach is compared with both global and local color correction approaches. Results show that our method outperforms the technique ranked first in a recent performance evaluation on this topic. Moreover, the proposed approach is computed in about one tenth of the time.

Abstract

Abstract