Reconstrucció 3D : comparativa d'algoritmes de correspondència

En aquest projecte s'analitzen dos algoritmes de correspondència entre imatges amb l'objectiu d'accelerar el procés de reconstrucció 3D mitjançant MVS. S'analitza tot el procés de reconstrucció i a partir d'un software existent es fa la comparació de l'algoritme SIFT i l'algoritme BRISK. A partir dels tests realitzats es conclou que el BRISK és més ràpid i millor per a una reconstrucció 3D.

Development of a system for 3D high-resolution seismic reflection profiling on lakes

A high-resolution three-dimensional (3D) seismic reflection system for small-scale targets in lacustrine settings has been developed. Its main characteristics include navigation and shot-triggering software that fires the seismic source at regular distance intervals (max. error of 0.25 m) with real-time control on navigation using differential GPS (Global Positioning System). Receiver positions are accurately calculated (error < 0.20 m) with the aid of GPS antennas attached to the end of each of three 24-channel streamers. Two telescopic booms hold the streamers at a distance of 7.5 m from each other. With a receiver spacing of 2.5 m, the bin dimension is 1.25 m in inline and 3.75 m in crossline direction. To test the system, we conducted a 3D survey of about 1 km(2) in Lake Geneva, Switzerland, over a complex fault zone. A 5-m shot spacing resulted in a nominal fold of 6. A double-chamber bubble-cancelling 15/15 in(3) air gun (40-650 Hz) operated at 80 bars and 1 m depth gave a signal penetration of 300 m below water bottom and a best vertical resolution of 1.1 m. Processing followed a conventional scheme, but had to be adapted to the high sampling rates, and our unconventional navigation data needed conversion to industry standards. The high-quality data enabled us to construct maps of seismic horizons and fault surfaces in three dimensions. The system proves to be well adapted to investigate complex structures by providing non-aliased images of reflectors with dips up to 30 degrees.

A High-Resolution 2D/3D Seismic Study of a Thrust Fault Zone in Lake Geneva, Switzerland

P130 A HIGH-RESOLUTION 2D/3D SEISMIC STUDY OF A THRUST FAULT ZONE IN LAKE GENEVA SWITZERLAND M. SCHEIDHAUER M. BERES D. DUPUY and F. MARILLIER Institute of Geophysics University of Lausanne 1015 Lausanne, Switzerland Summary A high-resolution three-dimensional (3D) seismic reflection survey has been conducted in Lake Geneva near the city of Lausanne Switzerland where the faulted molasse basement (Tertiary sandstones) is overlain by complex Quaternary sedimentary structures. Using a single 48-channel streamer an area of 1200 m x 600 m was surveyed in 10 days. With a 5-m shot spacing and a receiver spacing of 2.5 m in the inline direction and 7.5 m in the crossline direction, a 12-fold data coverage was achieved. A maximum penetration depth of ~150 m was achieved with a 15 cu. in. water gun operated at 140 bars. The multi-channel data allow the determination of an accurate velocity field for 3D processing, and they show particularly clean images of the fault zone and the overlying sediments in horizontal and vertical sections. In order to compare different sources, inline 55 was repeated with a 30/30 and a 15/15 cu. in. double-chamber air gun (Mini GI) operated at 100 and 80 bars, respectively. A maximum penetration depth of ~450 m was achieved with this source.

Feature Extraction from Whole-Sky Ground-Based Images for Cloud-Type Recognition

Several features that can be extracted from digital images of the sky and that can be useful for cloud-type classification of such images are presented. Some features are statistical measurements of image texture, some are based on the Fourier transform of the image and, finally, others are computed from the image where cloudy pixels are distinguished from clear-sky pixels. The use of the most suitable features in an automatic classification algorithm is also shown and discussed. Both the features and the classifier are developed over images taken by two different camera devices, namely, a total sky imager (TSI) and a whole sky imager (WSC), which are placed in two different areas of the world (Toowoomba, Australia; and Girona, Spain, respectively). The performance of the classifier is assessed by comparing its image classification with an a priori classification carried out by visual inspection of more than 200 images from each camera. The index of agreement is 76% when five different sky conditions are considered: clear, low cumuliform clouds, stratiform clouds (overcast), cirriform clouds, and mottled clouds (altocumulus, cirrocumulus). Discussion on the future directions of this research is also presented, regarding both the use of other features and the use of other classification techniques

Quantitative analysis of crystal/grain sizes and their distributions in 2D and 3D

We review methods to estimate the average crystal (grain) size and the crystal (grain) size distribution in solid rocks. Average grain sizes often provide the base for stress estimates or rheological calculations requiring the quantification of grain sizes in a rock's microstructure. The primary data for grain size data are either 1D (i.e. line intercept methods), 2D (area analysis) or 3D (e.g., computed tomography, serial sectioning). These data have been used for different data treatments over the years, whereas several studies assume a certain probability function (e.g., logarithm, square root) to calculate statistical parameters as the mean, median, mode or the skewness of a crystal size distribution. The finally calculated average grain sizes have to be compatible between the different grain size estimation approaches in order to be properly applied, for example, in paleo-piezometers or grain size sensitive flow laws. Such compatibility is tested for different data treatments using one- and two-dimensional measurements. We propose an empirical conversion matrix for different datasets. These conversion factors provide the option to make different datasets compatible with each other, although the primary calculations were obtained in different ways. In order to present an average grain size, we propose to use the area-weighted and volume-weighted mean in the case of unimodal grain size distributions, respectively, for 2D and 3D measurements. The shape of the crystal size distribution is important for studies of nucleation and growth of minerals. The shape of the crystal size distribution of garnet populations is compared between different 2D and 3D measurements, which are serial sectioning and computed tomography. The comparison of different direct measured 3D data; stereological data and direct presented 20 data show the problems of the quality of the smallest grain sizes and the overestimation of small grain sizes in stereological tools, depending on the type of CSD. (C) 2011 Published by Elsevier Ltd.

Reconeixement de gestos de la mà amb el sensor Kinect

El reconeixement dels gestos de la mà (HGR, Hand Gesture Recognition) és actualment un camp important de recerca degut a la varietat de situacions en les quals és necessari comunicar-se mitjançant signes, com pot ser la comunicació entre persones que utilitzen la llengua de signes i les que no. En aquest projecte es presenta un mètode de reconeixement de gestos de la mà a temps real utilitzant el sensor Kinect per Microsoft Xbox, implementat en un entorn Linux (Ubuntu) amb llenguatge de programació Python i utilitzant la llibreria de visió artifical OpenCV per a processar les dades sobre un ordinador portàtil convencional. Gràcies a la capacitat del sensor Kinect de capturar dades de profunditat d’una escena es poden determinar les posicions i trajectòries dels objectes en 3 dimensions, el que implica poder realitzar una anàlisi complerta a temps real d’una imatge o d’una seqüencia d’imatges. El procediment de reconeixement que es planteja es basa en la segmentació de la imatge per poder treballar únicament amb la mà, en la detecció dels contorns, per després obtenir l’envolupant convexa i els defectes convexos, que finalment han de servir per determinar el nombre de dits i concloure en la interpretació del gest; el resultat final és la transcripció del seu significat en una finestra que serveix d’interfície amb l’interlocutor. L’aplicació permet reconèixer els números del 0 al 5, ja que s’analitza únicament una mà, alguns gestos populars i algunes de les lletres de l’alfabet dactilològic de la llengua de signes catalana. El projecte és doncs, la porta d’entrada al camp del reconeixement de gestos i la base d’un futur sistema de reconeixement de la llengua de signes capaç de transcriure tant els signes dinàmics com l’alfabet dactilològic.

Trinocular System for 3D motion and Dense Stucture Estimation

The relief of the seafloor is an important source of data for many scientists. In this paper we present an optical system to deal with underwater 3D reconstruction. This system is formed by three cameras that take images synchronously in a constant frame rate scheme. We use the images taken by these cameras to compute dense 3D reconstructions. We use Bundle Adjustment to estimate the motion ofthe trinocular rig. Given the path followed by the system, we get a dense map of the observed scene by registering the different dense local reconstructions in a unique and bigger one

Identification of geometrically consistent interest points for 3D scene reconstruction

We propose an algorithm that extracts image features that are consistent with the 3D structure of the scene. The features can be robustly tracked over multiple views and serve as vertices of planar patches that suitably represent scene surfaces, while reducing the redundancy in the description of 3D shapes. In other words, the extracted features will off er good tracking properties while providing the basis for 3D reconstruction with minimum model complexity

Analysis of warm season thunderstorms using an object-oriented tracking method based on radar and total lightning data

Monitoring thunderstorms activity is an essential part of operational weather surveillance given their potential hazards, including lightning, hail, heavy rainfall, strong winds or even tornadoes. This study has two main objectives: firstly, the description of a methodology, based on radar and total lightning data to characterise thunderstorms in real-time; secondly, the application of this methodology to 66 thunderstorms that affected Catalonia (NE Spain) in the summer of 2006. An object-oriented tracking procedure is employed, where different observation data types generate four different types of objects (radar 1-km CAPPI reflectivity composites, radar reflectivity volumetric data, cloud-to-ground lightning data and intra-cloud lightning data). In the framework proposed, these objects are the building blocks of a higher level object, the thunderstorm. The methodology is demonstrated with a dataset of thunderstorms whose main characteristics, along the complete life cycle of the convective structures (development, maturity and dissipation), are described statistically. The development and dissipation stages present similar durations in most cases examined. On the contrary, the duration of the maturity phase is much more variable and related to the thunderstorm intensity, defined here in terms of lightning flash rate. Most of the activity of IC and CG flashes is registered in the maturity stage. In the development stage little CG flashes are observed (2% to 5%), while for the dissipation phase is possible to observe a few more CG flashes (10% to 15%). Additionally, a selection of thunderstorms is used to examine general life cycle patterns, obtained from the analysis of normalized (with respect to thunderstorm total duration and maximum value of variables considered) thunderstorm parameters. Among other findings, the study indicates that the normalized duration of the three stages of thunderstorm life cycle is similar in most thunderstorms, with the longest duration corresponding to the maturity stage (approximately 80% of the total time).

A filtering method to correct time-lapse 3D ERT data and improve imaging of natural aquifer dynamics

We have developed a processing methodology that allows crosshole ERT (electrical resistivity tomography) monitoring data to be used to derive temporal fluctuations of groundwater electrical resistivity and thereby characterize the dynamics of groundwater in a gravel aquifer as it is infiltrated by river water. Temporal variations of the raw ERT apparent-resistivity data were mainly sensitive to the resistivity (salinity), temperature and height of the groundwater, with the relative contributions of these effects depending on the time and the electrode configuration. To resolve the changes in groundwater resistivity, we first expressed fluctuations of temperature-detrended apparent-resistivity data as linear superpositions of (i) time series of riverwater-resistivity variations convolved with suitable filter functions and (ii) linear and quadratic representations of river-water-height variations multiplied by appropriate sensitivity factors; river-water height was determined to be a reliable proxy for groundwater height. Individual filter functions and sensitivity factors were obtained for each electrode configuration via deconvolution using a one month calibration period and then the predicted contributions related to changes in water height were removed prior to inversion of the temperature-detrended apparent-resistivity data. Applications of the filter functions and sensitivity factors accurately predicted the apparent-resistivity variations (the correlation coefficient was 0.98). Furthermore, the filtered ERT monitoring data and resultant time-lapse resistivity models correlated closely with independently measured groundwater electrical resistivity monitoring data and only weakly with the groundwater-height fluctuations. The inversion results based on the filtered ERT data also showed significantly less inversion artefacts than the raw data inversions. We observed resistivity increases of up to 10% and the arrival time peaks in the time-lapse resistivity models matched those in the groundwater resistivity monitoring data.

Reproducibility of 3D free-breathing magnetic resonance coronary vessel wall imaging.

AIMS: Although the coronary artery vessel wall can be imaged non-invasively using magnetic resonance imaging (MRI), the in vivo reproducibility of wall thickness measures has not been previously investigated. Using a refined magnetization preparation scheme, we sought to assess the reproducibility of three-dimensional (3D) free-breathing black-blood coronary MRI in vivo. METHODS AND RESULTS: MRI vessel wall scans parallel to the right coronary artery (RCA) were obtained in 18 healthy individuals (age range 25-43, six women), with no known history of coronary artery disease, using a 3D dual-inversion navigator-gated black-blood spiral imaging sequence. Vessel wall scans were repeated 1 month later in eight subjects. The visible vessel wall segment and the wall thickness were quantitatively assessed using a semi-automatic tool and the intra-observer, inter-observer, and inter-scan reproducibilities were determined. The average imaged length of the RCA vessel wall was 44.5+/-7 mm. The average wall thickness was 1.6+/-0.2 mm. There was a highly significant intra-observer (r=0.97), inter-observer (r=0.94), and inter-scan (r=0.90) correlation for wall thickness (all P<0.001). There was also a significant agreement for intra-observer, inter-observer, and inter-scan measurements on Bland-Altman analysis. The intra-class correlation coefficients for intra-observer (r=0.97), inter-observer (r=0.92), and inter-scan (r=0.86) analyses were also excellent. CONCLUSION: The use of black-blood free-breathing 3D MRI in conjunction with semi-automated analysis software allows for reproducible measurements of right coronary arterial vessel-wall thickness. This technique may be well-suited for non-invasive longitudinal studies of coronary atherosclerosis.