Processat de vídeo aplicat a la robòtica

Aquest projecte consisteix en acoblar una càmera web uEye en una de les extremitats d'un robot industrial (ABB Irc5). Aquest robot, es mourà en cas què es detecti una cara a través de la webcam. L'objectiu és intentar que la cara es situï sempre al mig de la imatge captada per la càmera. Podem dividir les tasques d'aquest projecte amb 4 fases diferents. A continuació expliquem les diferents etapes. La Càmera uEye s'encarrega de capturar imatges i passar-les a un PC utilitzant Python. La funció de Detecció de Cares ens indica la posició i dimensions de la cara. Tot seguit, la funció LabJack s'encarrega de canviar els valors digitals de les 5 senyals mitjançant programació amb Python. Per últim el Robot ABB interpreta el valor d'aquestes 5 senyals i es desplaça mitjançant petits increments amb l'objectiu de situar la cara al mig de la imatge capturada.

A Motion compensated filtering approach to remove sunlight flicker in shallow water images

A common problem in video surveys in very shallow waters is the presence of strong light fluctuations, due to sun light refraction. Refracted sunlight casts fast moving patterns, which can significantly degrade the quality of the acquired data. Motivated by the growing need to improve the quality of shallow water imagery, we propose a method to remove sunlight patterns in video sequences. The method exploits the fact that video sequences allow several observations of the same area of the sea floor, over time. It is based on computing the image difference between a given reference frame and the temporal median of a registered set of neighboring images. A key observation is that this difference will have two components with separable spectral content. One is related to the illumination field (lower spatial frequencies) and the other to the registration error (higher frequencies). The illumination field, recovered by lowpass filtering, is used to correct the reference image. In addition to removing the sunflickering patterns, an important advantage of the approach is the ability to preserve the sharpness in corrected image, even in the presence of registration inaccuracies. The effectiveness of the method is illustrated in image sets acquired under strong camera motion containing non-rigid benthic structures. The results testify the good performance and generality of the approach

Planar homography: accuracy analysis and applications

Projective homography sits at the heart of many problems in image registration. In addition to many methods for estimating the homography parameters (R.I. Hartley and A. Zisserman, 2000), analytical expressions to assess the accuracy of the transformation parameters have been proposed (A. Criminisi et al., 1999). We show that these expressions provide less accurate bounds than those based on the earlier results of Weng et al. (1989). The discrepancy becomes more critical in applications involving the integration of frame-to-frame homographies and their uncertainties, as in the reconstruction of terrain mosaics and the camera trajectory from flyover imagery. We demonstrate these issues through selected examples

Globally aligned photomosaic of the Lucky Strike hydrothermal vent field (Mid-Atlantic Ridge, 37°18.5’N): Release of georeferenced data, mosaic construction, and viewing software

We present a georeferenced photomosaic of the Lucky Strike hydrothermal vent field (Mid-Atlantic Ridge, 37°18’N). The photomosaic was generated from digital photographs acquired using the ARGO II seafloor imaging system during the 1996 LUSTRE cruise, which surveyed a ~1 km2 zone and provided a coverage of ~20% of the seafloor. The photomosaic has a pixel resolution of 15 mm and encloses the areas with known active hydrothermal venting. The final mosaic is generated after an optimization that includes the automatic detection of the same benthic features across different images (feature-matching), followed by a global alignment of images based on the vehicle navigation. We also provide software to construct mosaics from large sets of images for which georeferencing information exists (location, attitude, and altitude per image), to visualize them, and to extract data. Georeferencing information can be provided by the raw navigation data (collected during the survey) or result from the optimization obtained from imatge matching. Mosaics based solely on navigation can be readily generated by any user but the optimization and global alignment of the mosaic requires a case-by-case approach for which no universally software is available. The Lucky Strike photomosaics (optimized and navigated-only) are publicly available through the Marine Geoscience Data System (MGDS, http://www.marine-geo.org). The mosaic-generating and viewing software is available through the Computer Vision and Robotics Group Web page at the University of Girona (http://eia.udg.es/_rafa/mosaicviewer.html)

Small scale underwater change detection

In this paper, we present a method to deal with the constraints of the underwater medium for finding changes between sequences of underwater images. One of the main problems of underwater medium for automatically detecting changes is the low altitude of the camera when taking pictures. This emphasise the parallax effect between the images as they are not taken exactly at the same position. In order to solve this problem, we are geometrically registering the images together taking into account the relief of the scene

Creating large and accurate mosaics of the mid-atlantic ridge

We present an approach for creating image mosaics using navigation data consisting on 3D position estimates provided by sensors such as LBL available in deep water surveys. A central issue with acoustic 3D positioning is that the accuracy is far too low compositing the images within reasonable accuracy

A noninvasive method for measuring the velocity of diffuse hydrothermal flow by tracking moving refractive index anomalies

Diffuse flow velocimetry (DFV) is introduced as a new, noninvasive, optical technique for measuring the velocity of diffuse hydrothermal flow. The technique uses images of a motionless, random medium (e.g.,rocks) obtained through the lens of a moving refraction index anomaly (e.g., a hot upwelling). The method works in two stages. First, the changes in apparent background deformation are calculated using particle image velocimetry (PIV). The deformation vectors are determined by a cross correlation of pixel intensities across consecutive images. Second, the 2-D velocity field is calculated by cross correlating the deformation vectors between consecutive PIV calculations. The accuracy of the method is tested with laboratory and numerical experiments of a laminar, axisymmetric plume in fluids with both constant and temperaturedependent viscosity. Results show that average RMS errors are ∼5%–7% and are most accurate in regions of pervasive apparent background deformation which is commonly encountered in regions of diffuse hydrothermal flow. The method is applied to a 25 s video sequence of diffuse flow from a small fracture captured during the Bathyluck’09 cruise to the Lucky Strike hydrothermal field (September 2009). The velocities of the ∼10°C–15°C effluent reach ∼5.5 cm/s, in strong agreement with previous measurements of diffuse flow. DFV is found to be most accurate for approximately 2‐D flows where background objects have a small spatial scale, such as sand or gravel

Reconfigurable Architecture To Estimate The Motion Of An Underwater Vehicle

This work proposes a parallel architecture for a motion estimation algorithm. It is well known that image processing requires a huge amount of computation, mainly at low level processing where the algorithms are dealing with a great numbers of data-pixel. One of the solutions to estimate motions involves detection of the correspondences between two images. Due to its regular processing scheme, parallel implementation of correspondence problem can be an adequate approach to reduce the computation time. This work introduces parallel and real-time implementation of such low-level tasks to be carried out from the moment that the current image is acquired by the camera until the pairs of point-matchings are detected

Exemplar-based interpolation of sparsely sampled images

A nonlocal variational formulation for interpolating a sparsel sampled image is introduced in this paper. The proposed variational formulation, originally motivated by image inpainting problems, encouragesthe transfer of information between similar image patches, following the paradigm of exemplar-based methods. Contrary to the classical inpaintingproblem, no complete patches are available from the sparse imagesamples, and the patch similarity criterion has to be redefined as here proposed. Initial experimental results with the proposed framework, at very low sampling densities, are very encouraging. We also explore somedepartures from the variational setting, showing a remarkable ability to recover textures at low sampling densities.

Augmented state Kalman filtering for AUV navigation

Addresses the problem of estimating the motion of an autonomous underwater vehicle (AUV), while it constructs a visual map ("mosaic" image) of the ocean floor. The vehicle is equipped with a down-looking camera which is used to compute its motion with respect to the seafloor. As the mosaic increases in size, a systematic bias is introduced in the alignment of the images which form the mosaic. Therefore, this accumulative error produces a drift in the estimation of the position of the vehicle. When the arbitrary trajectory of the AUV crosses over itself, it is possible to reduce this propagation of image alignment errors within the mosaic. A Kalman filter with augmented state is proposed to optimally estimate both the visual map and the vehicle position

Motion estimation using higher-order statistics

The objective of this paper is to introduce a fourth-order cost function of the displaced frame difference (DFD) capable of estimatingmotion even for small regions or blocks. Using higher than second-orderstatistics is appropriate in case the image sequence is severely corruptedby additive Gaussian noise. Some results are presented and compared to those obtained from the mean kurtosis and the mean square error of the DFD.

Aplicació per a la predicció automàtica de la data d'adquisició d'una imatge

En aquesta memòria es presenta un projecte que té com a objectiu principal la creació d'una aplicació que, donada una imatge d'entrada, retorni en quina dècada va ser adquirida la imatge. Apart es pretén millorar els resultats obtinguts en altres estudis sobre la classificació d'imatges segons la seva data d'adquisició. En la memòria s'explica l'estudi realitzat anteriorment sobre el tema i quin ha estat el mètode escollit per millorar els seus resultats. També s'explica com hem creat l'aplicació i els passos que segueix l'aplicació en la seva execució.

Estudi per a una interfície d'ajuda a daltònics en la interpretació de mapes de metro

En aquest projecte fem un estudi de diferents mètodes per a la segmentació i extracció de línies de mapes de metro com a suport per a daltònics. Hem aplicat dos mètodes amb intervenció de l’usuari i cinc mètodes automàtics on fem servir K-means per a la segmentació de color i Hough per a l’extracció de línies. Dels mètodes amb intervenció obtenim millors resultats amb un mètode d’assignació aproximada del color, i entre els autoàatics tenim com a millor una solució ad-hoc sense paràmetres aplicada sobre l’espai RGB. D’acord amb els resultats experimentals, aquests mètodes ens permeten fer una bona segmentació i extracció de les línies de metro.

An embedded real-time red peach detection system based on an OV7670 camera, ARM Cortex-M4 processor and 3D Look-Up Tables

This work proposes the development of an embedded real-time fruit detection system for future automatic fruit harvesting. The proposed embedded system is based on an ARM Cortex-M4 (STM32F407VGT6) processor and an Omnivision OV7670 color camera. The future goal of this embedded vision system will be to control a robotized arm to automatically select and pick some fruit directly from the tree. The complete embedded system has been designed to be placed directly in the gripper tool of the future robotized harvesting arm. The embedded system will be able to perform real-time fruit detection and tracking by using a three-dimensional look-up-table (LUT) defined in the RGB color space and optimized for fruit picking. Additionally, two different methodologies for creating optimized 3D LUTs based on existing linear color models and fruit histograms were implemented in this work and compared for the case of red peaches. The resulting system is able to acquire general and zoomed orchard images and to update the relative tracking information of a red peach in the tree ten times per second.