Desarrollo de un sistema de estimación de mapas de profundidad densos a partir de secuencias reales de vídeo 3D.

Este proyecto fín de carrera describe el desarrollo de un sistema de estimación de mapas de profundidad densos a partir de secuencias reales de vídeo 3D. Está motivado por la necesidad de utilizar la información de profundidad de un vídeo estéreo para calcular las oclusiones en el módulo de inserción de objetos sintéticos interactivos desarrollado en el proyecto ImmersiveTV. En el receptor 3DTV, el sistema debe procesar en tiempo real secuencias estéreo de escenas reales en alta resolución con formato Side-by-Side. Se analizan las características del contenido para conocer los problemas a enfrentar. Obtener un mapa de profundidad denso mediante correspondencia estéreo (stereo matching) permite calcular las oclusiones del objeto sintético con la escena. No es necesario que el valor de disparidad asignado a cada píxel sea preciso, basta con distinguir los distintos planos de profundidad ya que se trabaja con distancias relativas. La correspondencia estéreo exige que las dos vistas de entrada estén alineadas. Primero se comprueba si se deben rectificar y se realiza un repaso teórico de calibración y rectificación, resumiendo algunos métodos a considerar en la resolución del problema. Para estimar la profundidad, se revisan técnicas de correspondencia estéreo densa habituales, seleccionando un conjunto de implementaciones con el fin de valorar cuáles son adecuadas para resolver el problema, incluyendo técnicas locales, globales y semiglobales, algunas sobre CPU y otras para GPU; modificando algunas para soportar valores negativos de disparidad. No disponer de ground truth de los mapas de disparidad del contenido real supone un reto que obliga a buscar métodos indirectos de comparación de resultados. Para una evaluación objetiva, se han revisado trabajos relacionados con la comparación de técnicas de correspondencia y entornos de evaluación existentes. Se considera el mapa de disparidad como error de predicción entre vistas desplazadas. A partir de la vista derecha y la disparidad de cada píxel, puede reconstruirse la vista izquierda y, comparando la imagen reconstruida con la original, se calculan estadísticas de error y las tasas de píxeles con disparidad inválida y errónea. Además, hay que tener en cuenta la eficiencia de los algoritmos midiendo la tasa de cuadros por segundo que pueden procesar. Observando los resultados, atendiendo a los criterios de maximización de PSNR y minimización de la tasa de píxeles incorrectos, se puede elegir el algoritmo con mejor comportamiento. Como resultado, se ha implementado una herramienta que integra el sistema de estimación de mapas de disparidad y la utilidad de evaluación de resultados. Trabaja sobre una imagen, una secuencia o un vídeo estereoscópico. Para realizar la correspondencia, permite escoger entre un conjunto de algoritmos que han sido adaptados o modificados para soportar valores negativos de disparidad. Para la evaluación, se ha implementado la reconstrucción de la vista de referencia y la comparación con la original mediante el cálculo de la RMS y PSNR, como medidas de error, además de las tasas de píxeles inválidos e incorrectos y de la eficiencia en cuadros por segundo. Finalmente, se puede guardar las imágenes (o vídeos) generados como resultado, junto con un archivo de texto en formato csv con las estadísticas para su posterior comparación.

Mitigating congestion by integrating time forecasting and realtime information aggregation in cellular networks

An iterative travel time forecasting scheme, named the Advanced Multilane Prediction based Real-time Fastest Path (AMPRFP) algorithm, is presented in this dissertation. This scheme is derived from the conventional kernel estimator based prediction model by the association of real-time nonlinear impacts that caused by neighboring arcs’ traffic patterns with the historical traffic behaviors. The AMPRFP algorithm is evaluated by prediction of the travel time of congested arcs in the urban area of Jacksonville City. Experiment results illustrate that the proposed scheme is able to significantly reduce both the relative mean error (RME) and the root-mean-squared error (RMSE) of the predicted travel time. To obtain high quality real-time traffic information, which is essential to the performance of the AMPRFP algorithm, a data clean scheme enhanced empirical learning (DCSEEL) algorithm is also introduced. This novel method investigates the correlation between distance and direction in the geometrical map, which is not considered in existing fingerprint localization methods. Specifically, empirical learning methods are applied to minimize the error that exists in the estimated distance. A direction filter is developed to clean joints that have negative influence to the localization accuracy. Synthetic experiments in urban, suburban and rural environments are designed to evaluate the performance of DCSEEL algorithm in determining the cellular probe’s position. The results show that the cellular probe’s localization accuracy can be notably improved by the DCSEEL algorithm. Additionally, a new fast correlation technique for overcoming the time efficiency problem of the existing correlation algorithm based floating car data (FCD) technique is developed. The matching process is transformed into a 1-dimensional (1-D) curve matching problem and the Fast Normalized Cross-Correlation (FNCC) algorithm is introduced to supersede the Pearson product Moment Correlation Co-efficient (PMCC) algorithm in order to achieve the real-time requirement of the FCD method. The fast correlation technique shows a significant improvement in reducing the computational cost without affecting the accuracy of the matching process.

On precise three-dimensional environment modeling via UAV-based photogrammetric systems

Abstract : Images acquired from unmanned aerial vehicles (UAVs) can provide data with unprecedented spatial and temporal resolution for three-dimensional (3D) modeling. Solutions developed for this purpose are mainly operating based on photogrammetry concepts, namely UAV-Photogrammetry Systems (UAV-PS). Such systems are used in applications where both geospatial and visual information of the environment is required. These applications include, but are not limited to, natural resource management such as precision agriculture, military and police-related services such as traffic-law enforcement, precision engineering such as infrastructure inspection, and health services such as epidemic emergency management. UAV-photogrammetry systems can be differentiated based on their spatial characteristics in terms of accuracy and resolution. That is some applications, such as precision engineering, require high-resolution and high-accuracy information of the environment (e.g. 3D modeling with less than one centimeter accuracy and resolution). In other applications, lower levels of accuracy might be sufficient, (e.g. wildlife management needing few decimeters of resolution). However, even in those applications, the specific characteristics of UAV-PSs should be well considered in the steps of both system development and application in order to yield satisfying results. In this regard, this thesis presents a comprehensive review of the applications of unmanned aerial imagery, where the objective was to determine the challenges that remote-sensing applications of UAV systems currently face. This review also allowed recognizing the specific characteristics and requirements of UAV-PSs, which are mostly ignored or not thoroughly assessed in recent studies. Accordingly, the focus of the first part of this thesis is on exploring the methodological and experimental aspects of implementing a UAV-PS. The developed system was extensively evaluated for precise modeling of an open-pit gravel mine and performing volumetric-change measurements. This application was selected for two main reasons. Firstly, this case study provided a challenging environment for 3D modeling, in terms of scale changes, terrain relief variations as well as structure and texture diversities. Secondly, open-pit-mine monitoring demands high levels of accuracy, which justifies our efforts to improve the developed UAV-PS to its maximum capacities. The hardware of the system consisted of an electric-powered helicopter, a high-resolution digital camera, and an inertial navigation system. The software of the system included the in-house programs specifically designed for camera calibration, platform calibration, system integration, onboard data acquisition, flight planning and ground control point (GCP) detection. The detailed features of the system are discussed in the thesis, and solutions are proposed in order to enhance the system and its photogrammetric outputs. The accuracy of the results was evaluated under various mapping conditions, including direct georeferencing and indirect georeferencing with different numbers, distributions and types of ground control points. Additionally, the effects of imaging configuration and network stability on modeling accuracy were assessed. The second part of this thesis concentrates on improving the techniques of sparse and dense reconstruction. The proposed solutions are alternatives to traditional aerial photogrammetry techniques, properly adapted to specific characteristics of unmanned, low-altitude imagery. Firstly, a method was developed for robust sparse matching and epipolar-geometry estimation. The main achievement of this method was its capacity to handle a very high percentage of outliers (errors among corresponding points) with remarkable computational efficiency (compared to the state-of-the-art techniques). Secondly, a block bundle adjustment (BBA) strategy was proposed based on the integration of intrinsic camera calibration parameters as pseudo-observations to Gauss-Helmert model. The principal advantage of this strategy was controlling the adverse effect of unstable imaging networks and noisy image observations on the accuracy of self-calibration. The sparse implementation of this strategy was also performed, which allowed its application to data sets containing a lot of tie points. Finally, the concepts of intrinsic curves were revisited for dense stereo matching. The proposed technique could achieve a high level of accuracy and efficiency by searching only through a small fraction of the whole disparity search space as well as internally handling occlusions and matching ambiguities. These photogrammetric solutions were extensively tested using synthetic data, close-range images and the images acquired from the gravel-pit mine. Achieving absolute 3D mapping accuracy of 11±7 mm illustrated the success of this system for high-precision modeling of the environment.

Evaluation of confidence-driven cost aggregation strategies

In this thesis I describe eight new stereo matching algorithms that perform the cost-aggregation step using a guided filter with a confidence map as guidance image, and share the structure of a linear stereo matching algorithm. The results of the execution of the proposed algorithms on four pictures from the Middlebury dataset are shown as well. Finally, based on these results, a ranking of the proposed algorithms is presented.

Stereo image matching using wavelet scale-space representation

A multi-resolution technique for matching a stereo pair of images based on translation invariant discrete multi-wavelet transform is presented. The technique uses the well known coarse to fine strategy, involving the calculation of matching points at the coarsest level with consequent refinement up to the finest level. Vector coefficients of the wavelet transform modulus are used as matching features, where modulus maxima defines the shift invariant high-level features (multiscale edges) with phase pointing to the normal of the feature surface. The technique addresses the estimation of optimal corresponding points and the corresponding 2D disparity maps. Illuminative variation that can exist between the perspective views of the same scene is controlled using scale normalization at each decomposition level by dividing the details space coefficients with approximation space and then using normalized correlation. The problem of ambiguity, explicitly, and occlusion, implicitly, is addressed by using a geometric topological refinement procedure and symbolic tagging.

Fast and robust stereo object recognition for spheres

The paper presents a fast and robust stereo object recognition method. The method is currently unable to identify the rotation of objects. This makes it very good at locating spheres which are rotationally independent. Approximate methods for located non-spherical objects have been developed. Fundamental to the method is that the correspondence problem is solved using information about the dimensions of the object being located. This is in contrast to previous stereo object recognition systems where the scene is first reconstructed by point matching techniques. The method is suitable for real-time application on low-power devices.

A Taxonomy of image matching techniques for stereo vision

Stereo vision is a method of depth perception, in which depth information is inferred from two (or more) images of a scene, taken from different perspectives. Applications of stereo vision include aerial photogrammetry, autonomous vehicle guidance, robotics, industrial automation and stereomicroscopy. A key issue in stereo vision is that of image matching, or identifying corresponding points in a stereo pair. The difference in the positions of corresponding points in image coordinates is termed the parallax or disparity. When the orientation of the two cameras is known, corresponding points may be projected back to find the location of the original object point in world coordinates. Matching techniques are typically categorised according to the nature of the matching primitives they use and the matching strategy they employ. This report provides a detailed taxonomy of image matching techniques, including area based, transform based, feature based, phase based, hybrid, relaxation based, dynamic programming and object space methods. A number of area based matching metrics as well as the rank and census transforms were implemented, in order to investigate their suitability for a real-time stereo sensor for mining automation applications. The requirements of this sensor were speed, robustness, and the ability to produce a dense depth map. The Sum of Absolute Differences matching metric was the least computationally expensive; however, this metric was the most sensitive to radiometric distortion. Metrics such as the Zero Mean Sum of Absolute Differences and Normalised Cross Correlation were the most robust to this type of distortion but introduced additional computational complexity. The rank and census transforms were found to be robust to radiometric distortion, in addition to having low computational complexity. They are therefore prime candidates for a matching algorithm for a stereo sensor for real-time mining applications. A number of issues came to light during this investigation which may merit further work. These include devising a means to evaluate and compare disparity results of different matching algorithms, and finding a method of assigning a level of confidence to a match. Another issue of interest is the possibility of statistically combining the results of different matching algorithms, in order to improve robustness.

A robust solution to the stereo-vision-based simultaneous localization and mapping problem with steady and moving landmarks

The problem of visual simultaneous localization and mapping (SLAM) is examined in this paper using recently developed ideas and algorithms from modern robust control and estimation theory. A nonlinear model for a stereo-vision-based sensor is derived that leads to nonlinear measurements of the landmark coordinates along with optical flow-based measurements of the relative robot-landmark velocity. Using a novel analytical measurement transformation, the nonlinear SLAM problem is converted into the linear domain and solved using a robust linear filter. Actually, the linear filter is guaranteed stable and the SLAM state estimation error is bounded within an ellipsoidal set. A mathematically rigorous stability proof is given that holds true even when the landmarks move in accordance with an unknown control input. No similar results are available for the commonly employed extended Kalman filter, which is known to exhibit divergence and inconsistency characteristics in practice. A number of illustrative examples are given using both simulated and real vision data that further validate the proposed method.

Concept of Matching Parallelepiped and its use in the correspondence problem

In this paper, the concept of Matching Parallelepiped (MP) is presented. It is shown that the volume of the MP can be used as an additional measure of `distance' between a pair of candidate points in a matching algorithm by Relaxation Labeling (RL). The volume of the MP is related with the Epipolar Geometry and the use of this measure works as an epipolar constraint in a RL process, decreasing the efforts in the matching algorithm since it is not necessary to explicitly determine the equations of the epipolar lines and to compute the distance of a candidate point to each epipolar line. As at the beginning of the process the Relative Orientation (RO) parameters are unknown, a initial matching based on gradient, intensities and correlation is obtained. Based on this set of labeled points the RO is determined and the epipolar constraint included in the algorithm. The obtained results shown that the proposed approach is suitable to determine feature-point matching with simultaneous estimation of camera orientation parameters even for the cases where the pair of optical axes are not parallel.

Variational Stereo Imaging of Oceanic Waves with Statistical Constraints

An image processing observational technique for the stereoscopic reconstruction of the wave form of oceanic sea states is developed. The technique incorporates the enforcement of any given statistical wave law modeling the quasi Gaussianity of oceanic waves observed in nature. The problem is posed in a variational optimization framework, where the desired wave form is obtained as the minimizer of a cost functional that combines image observations, smoothness priors and a weak statistical constraint. The minimizer is obtained combining gradient descent and multigrid methods on the necessary optimality equations of the cost functional. Robust photometric error criteria and a spatial intensity compensation model are also developed to improve the performance of the presented image matching strategy. The weak statistical constraint is thoroughly evaluated in combination with other elements presented to reconstruct and enforce constraints on experimental stereo data, demonstrating the improvement in the estimation of the observed ocean surface.

Simulation in the supply chain context:matching the simulation tool to the problem

The supply chain can be a source of competitive advantage for the firm. Simulation is an effective tool for investigating supply chain problems. The three main simulation approaches in the supply chain context are System Dynamics (SD), Discrete Event Simulation (DES) and Agent Based Modelling (ABM). A sample from the literature suggests that whilst SD and ABM have been used to address strategic and planning problems, DES has mainly been used on planning and operational problems., A review of received wisdom suggests that historically, driven by custom and practice, certain simulation techniques have been focused on certain problem types. A theoretical review of the techniques, however, suggests that the scope of their application should be much wider and that supply chain practitioners could benefit from applying them in this broader way.