Bretar F.,Normandie Center |
Bretar F.,Laboratoire Of Geomatique Appliquee |
Arab-Sedze M.,University Paris Diderot |
Arab-Sedze M.,Laboratoire MATIS |
And 4 more authors.
Remote Sensing of Environment | Year: 2013
We present a rapid in situ photogrammetric method to characterize surface roughness by taking overlapping photographs of a scene. The method uses a single digital camera to create a high-resolution digital terrain model (pixel size of ~1.32mm) by means of a free open-source stereovision software. It is based on an auto-calibration process, which calculates the 3D geometry of the images, and an efficient multi-image correlation algorithm. The method is successfully applied to four different volcanic surfaces-namely, a'a lava flows, pahoehoe lava flows, slabby pahoehoe lava flows, and lapilli deposits. These surfaces were sampled in the Piton de la Fournaise volcano (Reunion Island) in October, 2011, and displayed various terrain roughnesses. Our in situ measurements allow deriving digital terrain models that reproduce the millimeter-scale height variations of the surfaces over about 12m2. Five parameters characterizing surface topography are derived along unidirectional profiles: the root-mean-square height (ξ), the correlation length (Lc), the ratio Zs=ξ2/Lc, the tortuosity index (τ), and the fractal dimension (D). Anisotropy in the surface roughness has been first investigated using 1-m-long profiles circularly arranged around a central point. The results show that Lc, Zs and D effectively catch preferential directions in the structure of bare surfaces. Secondly, we studied the variation of these parameters as a function of the profile length by drawing random profiles from 1 to 12m in length. We verified that ξ and Lc increase with the profile length and, therefore, are not appropriate to characterize surface roughness variation. We conclude that Zs and D are better suited to extract roughness information for multiple eruptive terrains with complex surface texture. © 2013 Elsevier Inc.
Filho C.A.F.,Federal University of Minas Gerais |
Arajujo A.D.A.,Federal University of Minas Gerais |
Crucianu M.,Cnam CEDRIC |
Gouet-Brunet V.,Cnam CEDRIC |
Gouet-Brunet V.,Laboratoire MATIS
Brazilian Symposium of Computer Graphic and Image Processing | Year: 2013
Among various image retrieval approaches, the use of sketches lets one express a precise visual query with simple and widespread means. The challenge consists in finding a content representation that allows you to effectively compare sketches and images, while supporting efficient retrieval in order to make the system scalable. We put forward a sketch-based image retrieval solution where sketches and natural image contours are represented and compared in the wavelet domain. The relevant information regarding query sketches and image content has, thus, a compact representation that can be readily employed by an efficient index for retrieval by similarity. Furthermore, with this solution, the balance between effectiveness and efficiency can be easily modified in order to adapt to the available resources. A comparative evaluation with a state-of-the-art method on the Paris dataset and a subset with 535K images of the Image Net dataset shows that our solution can preserve effectiveness while being more than one order of magnitude faster. © 2013 IEEE.
Tournaire O.,Laboratoire MATIS |
Bredif M.,Laboratoire MATIS |
Boldo D.,Laboratoire MATIS |
Durupt M.,Laboratoire MATIS
ISPRS Journal of Photogrammetry and Remote Sensing | Year: 2010
In the past two decades, building detection and reconstruction from remotely sensed data has been an active research topic in the photogrammetric and remote sensing communities. Recently, effective high level approaches have been developed, i.e., the ones involving the minimization of an energetic formulation. Yet, their efficiency has to be balanced by the amount of processing power required to obtain good results. In this paper, we introduce an original energetic model for building footprint extraction from high resolution digital elevation models (≤1 m) in urban areas. Our goal is to formulate the energy in an efficient way, easy to parametrize and fast to compute, in order to get an effective process still providing good results. Our work is based on stochastic geometry, and in particular on marked point processes of rectangles. We therefore try to obtain a reliable object configuration described by a collection of rectangular building footprints. To do so, an energy function made up of two terms is defined: the first term measures the adequacy of the objects with respect to the data and the second one has the ability to favour or penalize some footprint configurations based on prior knowledge (alignment, overlapping, ...). To minimize the global energy, we use a Reversible Jump Monte Carlo Markov Chain (RJMCMC) sampler coupled with a simulated annealing algorithm, leading to an optimal configuration of objects. Various results from different areas and resolutions are presented and evaluated. Our work is also compared with an already existing methodology based on the same mathematical framework that uses a much more complex energy function. We show how we obtain similarly good results with a high computational efficiency (between 50 and 100 times faster) using a simplified energy that requires a single data-independent parameter, compared to more than 20 inter-related and hard-to-tune parameters. © 2010 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS).