Издательство InTech, 2008, -380 pp.Computer vision has gained enormous research interest over the last two decades with exponentially growing focus on stereo vision. Stereo vision has long been studied and lot of research involving new discoveries, techniques and applications has been reported and published over the years. Although, these published texts serve as fine introductions and references to the core mathematical ideas, however, cannot hope to keep pace with the vast and diverse outpouring of new research activities. In contrast, this volume has accumulated the most recent advances and trends of stereo vision research from around the globe. The goal of this book is to provide an insight of the current research trends and advances in the field of stereo vision. Furthermore, to provide a particularly good way for experts in one aspect of the field to learn about advances made by their colleagues with different research interests. It is quite understandable that visual information in 3D view possesses more information about the objects in the scene than its counterpart 2D view. During the image formation process of the cameras, explicit depth information about the scenes is lost. In many applications, such as industrial assembly and inspection, robot obstacle detection and path planning, autonomous vehicle navigation of unfamiliar environments, image based object modelling, surveillance and security, medical image analysis, and human-computer interaction, one of the most critical tasks is the recovery and estimation of depth information. Therefore, depth information has to be inferred implicitly from the 2D views of the scenes. There are a number of techniques available in the literature for depth estimation; however the most widely used techniques are based on stereo vision. Stereo vision techniques are mainly inspired by the human visual system, where two perspective views of eyes lead to slight offset of objects (disparities) in the two monocular views, leading to the phenomenon of depth perception. In spite of the fact that research on stereo vision spans over almost two decades; the performance of the stereo vision systems could not be compared with the human visual system. Stereo vision has long been studied and a number of techniques have been reported. Depth by stereo is achieved by estimating the pixel correspondences of similar features in the stereo perspective views originated from the same 3D scene. However, finding correct corresponding points is still an ill posed problem. Some of the factors that make the correspondence problem difficult are its inherent ambiguity, occlusions, photometric, radial and geometric distortions. A number of algorithms have been proposed to address some of the aforementioned problems in stereo vision however it, relatively, is still an open problem. To address aforementioned issues, the presented book consists of topics from theoretical aspects to the applications of stereo vision. The book consists of 20 chapters, each highlighting different aspects of the research in stereo vision. The book can be categorized into the spectrum of three broad interests that includes the theoretical aspects of the stereo vision, algorithm development for robust disparity (consequently depth) estimation and the applications of stereo vision in different domain. Generally, understanding theoretical aspects and the algorithm development in solving for the robust solutions goes hand in hand. Similarly, the algorithm development and the relevant applications are also tightly coupled as generally algorithms are customized to achieve optimum performance for specific applications which is hard to achieve, otherwise. For instance; chapters 18, 9, 20, 15, 19, 2, 17, 1 and 7 fall into the category of first two interests that involves the discussion of theoretical aspects as well as the development of algorithms in reference to stereo vision. Particularly, chapter 20 describes Graph Cut whereas chapter 2 describes multiresolution analysis based algorithms that fall into two famous classes of stereo vision algorithms, i.e. global optimization algorithm and multi-resolution based hierarchical algorithms, respectively. Furthermore, chapters 6, 8, 10, 11-14 and 16 present applications of stereo vision and current trends. The applications that have been covered in this book includes driver assistance system, robotic bin-picking, human computer interaction (HMI), head detection and tracking, 3D underwater mosaicking, moving target tracking, robot navigation, pose estimation, smart homes and 3D avatar communication. Chapters 5 and 14 present hardware based approaches regarding the development of stereo vision systems that has direct commercial applications. In summary this book comprehensively covers almost all aspects of stereo vision and highlights the current knowledge trends as well as corresponding solutions. In addition reader can find topics from defining knowledge gaps to the state of the art algorithms as well as current application trends of stereo vision to the development of intelligent hardware modules and smart cameras. It would not be an exaggeration if this book is considered to be one of the most comprehensive books published in reference to the current research in the field of stereo vision. Research topics covered in this book makes it equally essential and important for students and early career researchers as well as senior academics linked with computer vision.Calibration and Sensitivity Analysis of a Stereo Vision-Based Driver Assistance System. Stereo Correspondence Estimation based on Wavelets and Multiwavelets Analysis. Stereo Vision for Unrestricted Human-Computer Interaction. A Systems Engineering Approach to Robotic Bin Picking. Stereo Vision in Smart Camera Networks. Grayscale Correlation based 3D Model Fitting for Occupant Head Detection and Tracking. A Performance Review of 3D TOF Vision Systems in Comparison to Stereo Vision Systems. Construction of an Intelligent Room using Distributed Camera System. The Integrated Active Stereoscopic Vision Theory, Integration and Application. Arrangement of a Multi Stereo Visual Sensor System for a Human Activities Space. A Stereo Vision Framework for 3-D Underwater Mosaicking. Moving Target Tracking of Omnidirectional Robot with Stereo Cameras. High School Educational Program Using a Simple and Compact Stereo Vision Robot. A Sensor for Urban Driving Assistance Systems Based on Dense Stereovision. Stereo Algorithm with Reaction-Diffusion Equations. Point Cloud Streaming for 3D Avatar Communication. A Self Navigation Technique using Stereovision Analysis. Real Time Stereo Image Registration for Planar Structure and 3D Sensor Pose Estimation. Robust Visual Correspondence: Theory and Applications. Usinig Optical Flow as an Additional Constraint for Solving the Correspondence Problem in Binocular Stereopsis. Stereo Matching and Graph Cuts.
Чтобы скачать этот файл зарегистрируйтесь и/или войдите на сайт используя форму сверху.