Active binocular vision systems are powerful tools in machine vision. With a virtually unlimited field-of-view they have access to huge amounts of information, yet are able to confine their resources to specific regions of interest. Since they can dynamically interact with the environment, they are able to successfully address problems that are ill-posed to passive systems. A primary goal of an active binocular vision systems is to ascertain depth information. Since they employ two cameras and are able to sample a scene from two distinct vantage points, they are well suited for such a task. The depth recovery process is composed of two interrelated components: image registration and sampling. Image registration is the process of determining corresponding points between the stereo images. Once points in the images have been matched, 3D information can be recovered via triangulation. Image sampling determines how the image is discretized and represented. Image registration and sampling are highly interdependent. The choice of sampling scheme can profoundly impact the accuracy and complexity of the registrations process. In many situations, particular registration algorithms are simply incompatible with some sampling schemes. In this dissertation we meticulously address both registration and sampling in the context of stereopis for active binocular vision systems. Throughout the development of this work, contributions in each area are addressed with an eye toward their eventual integration into a cohesive registration procedure appropriate for active binocular vision systems. The actual synthesis is a daunting task that is beyond the scope of this single dissertation. The focus of this work is to assiduously analyze both registration and sampling, establishing a solid foundation for their future aggregation. One of the most successful approaches to image registration is phase-differencing. Phase-differencing algorithms provide a fast, powerful means for depth recovery. Unfortunately, phase-differencing techniques suffer from two significant impediments: phase nonlinearities and neglect of multispectral information. This dissertation uses the amenable properties of white noise images to analytically quantify the behavior of phase in these regions of phase nonlinearity. The improved understanding gained from this analysis enables us to create a new, more effective method for identifying these regions based on the second derivative of phase. We also suggest a novel approach that combines our method of nonlinear phase detection with strategies of both phase-differencing and local correlation. This hybrid approach retains the advantageous properties of phase-differencing while incorporating the multispectral aspects of local correlation. This task of registration is greatly simplified if the camera geometry is known and the search for corresponding points can be restricted to epipolar lines. Unfortunately, computation of epipolar lines for an active system requires calibration which can be both highly complex and inaccurate. While it is possible to register images without calibration information, such unconstrained algorithms are usually time consuming and prone to error. In this dissertation we propose compromise. Even without the instantaneous knowledge of the system geometry, we can restrict the region of correspondence by imposing limits on the possible range of configurations, and as a result, confine our search for matching points to what we refer to as epipolar spaces. For each point in one image, we define the corresponding epipolar space in the other image as the union of all associated epipolar lines over all possible system geometries. Epipolar spaces eliminate the need for calibration at the cost of an increased search region. Since the average size of a search space is directly related to the accuracy and efficiency of any registration algorithm, it is essential to mitigate the increase. The major contribution of this dissertation is the derivation of an optimal nonuniform sampling that minimizes the average area per epipolar space. / text

No description available.

Thesis (Ph.D.)--University of Alabama at Birmingham, 2010. / Title from PDF title page (viewed on July 8, 2010). Includes bibliographical references (p.169-183).

Thesis (Ph. D.)--University of California, Santa Cruz, 1989. / Typescript. Includes bibliographical references.

The Pedestrian Detection System (PDS) has become a significant area of research designed to protect pedestrians. Despite the huge number of research work, the most current PDSs are designed to detect pedestrians without knowing their distances from cars. In fact, a priori knowledge of the distance between a car and pedestrian allows this system to make the appropriate decision in order to avoid collisions. Typical methods of distance measurement require additional equipment (e.g., Radars) which, unfortunately, cannot identify objects. Moreover, traditional stereo-vision methods have poor precision in long-range conditions. In this thesis, we use the keypoint-based feature extraction method to generate the parallax in a binocular vision system in order to measure a detectable object; this is used instead of a disparity map. Our method enhances the tolerance to instability of a moving vehicle; and, it also enables binocular measurement systems to be equipped with a zoom lens and to have greater distance between cameras. In addition, we designed a crossover re-detection and tracking method in order to reinforce the robustness of the system (one camera helps the other reduce detection errors). Our system is able to measure the distance between cars and pedestrians; and, it can also be used efficiently to measure the distance between cars and other objects such as Traffic signs or animals. Through a real word experiment, the system shows a 7.5% margin of error in outdoor and long-range conditions.

yes / Binocularity in the human visual system poses two interesting and extremely challenging questions. The first, and perhaps most obvious stems from the singularity of perception even though the neural images we see originate as two separate images in the right and left eyes. Mechanistically we can ask how and where do we convert two images into one? The second question is more of a “why” question. By converting lateral eyes with their inherent panoramic visual field into frontal eyes with overlapping binocular visual fields, primates have developed an extremely large blind region (the half of the world behind us). We generally accept that this sacrifice in visual field size was driven by the potential benefit of extracting information about the 3rd dimension from overlapping right and left eye visual fields. For some people, both of these core processes of binocularity fail: a single fused binocular image is not achieved (when diplopia or suppression is present), and the ability to accurately represent the 3rd dimension is lost (stereo-blindness). In addition to these failures in the core functions of the human binocular system, early imbalances in the quality of right and left eye neural images (e.g. due to anisometropia, monocular deprivation, and/or strabismus), can precipitate profound neurological changes at a cortical level which can lead to serious vision loss in one eye (amblyopia). Caring for patients with malfunctioning binocular visual systems is a core therapeutic responsibility of the eye care professions (optometry, ophthalmology and orthoptics) and significant advances in patient care and subsequent visual outcomes will be gained from a deeper understanding of how the human brain accomplishes full binocular integration. This feature issue on binocular vision brings together original articles and reviews from leading groups of neuroscientists, psychophysicists and clinical scientists from around the world who embrace the multidisciplinary nature of this topic. Our authors have taken on the big issues facing the research community tasked with understanding how binocular vision is meant to work, how it fails, and how to better treat those with compromised binocularity. These studies address deep issues about how the human brain functions and how it fails, as well as how it can be altered by therapy.

In the past decade, considerable attention has been paid to examination of the impact of amblyopia, and strabismus, upon the lives of the individuals. Although an extensive amount of literature exists regarding amblyopia and its associated visual defects, little is known about the contribution of the amblyopic eye in the habitual viewing condition (i.e. both eyes viewing). The purpose of these studies was to determine whether amblyopes are disadvantaged in the performance of tasks under habitual viewing conditions, highlighting any functional differences which may exist as a consequence of amblyopia. Secondly, the work aimed to investigate whether the amblyopic eye contributes to the habitual performance of these tasks. A simple light detection task, in a dichoptic arrangement based upon blue/yellow stimuli viewed through yellow filters, was used to investigate the above two aims and investigate the degree of interocular suppression in amblyopic participants. Using a 3D motion analysis system performance was assessed for an obstacle crossing task (adaptive gait) and a task of reaching for and grasping of an isolated object and in a ¿cluttered¿ environment. Fine motor skills were assessed in a threading a needle task. On the whole it was found that amblyopes are not disadvantaged under habitual viewing conditions, and in cases where differences were found to exist this appeared to be in tasks requiring speed and accuracy. Consistently across all studies it was found that the amblyopic eye contributed in a positive manner, thus, as in visual normals, two eyes are better than one.

No description available.

No description available.

Binocular camera fixation refers to the control of camera movements so that the optical axes of two cameras converge at a single point of interest in a three-dimensional scene. This thesis describes the theory and high-level algorithms for a procedure that can locate fixation points in an image pair. The binocular fixation procedure which has been developed here combats several difficult fixation problems that are hard to overcome with traditional approaches. Examples of this are objects lacking in visual texture, scenes with occluding conditions, repeating patterns, and objects which are steeply slanted with respect to the image planes. In order to deal with low-contrast image regions, the system is capable of detecting nearby salient points and shifting attention to a new salient target. Based on local image information, the system adaptively determines an initial window size using an Autoregressive-Moving Average (ARMA) modeling technique. Area-based feature matching is performed using normalized cross-covariance within a multiresolution image hierarchy. Gaussian low-pass filters of increasing spatial resolution are used to construct the hierarchy, and matching is performed in coarse-to-fine order. Finally, affine image warping is used in some cases to reduce the effects of perspective distortion so that area based image comparison is improved. The method presented here can serve as the foundation of a stereo tracking system, and has potential applications in such areas as autonomous vehicle navigation, industrial process control, and visually guided robot manipulation. / Master of Science