High-order Spatial and Spatiotemporal Image Statistics and Visual Processing PDF Download
Are you looking for read ebook online? Search for your book and save it on your Kindle device, PC, phones or tablets. Download High-order Spatial and Spatiotemporal Image Statistics and Visual Processing PDF full book. Access full book title High-order Spatial and Spatiotemporal Image Statistics and Visual Processing by Qin Hu. Download full books in PDF and EPUB format.
Author: Qin Hu Publisher: ISBN: Category : Languages : en Pages : 202
Book Description
This thesis explores the relationship of image statistics to visual processing by considering two important aspects: motion processing and static images. Detection of movement is one of the most fundamental tasks performed by our visual system. The essence of motion is spatiotemporal correlation, but how these correlations are processed biologically is not yet fully known. To probe the computations underlying motion perception, we created a new class of motion stimuli, "glider motion," characterized by their third- and fourth-order spatiotemporal correlations. The direction of glider motion cannot be detected by current models for the neural computation of spatiotemporal correlations. Nevertheless, glider stimuli reliably produced a directional motion percept in humans. This implies that the current models for the computations underlying motion processing require modification, as we discuss. To determine where the calculation of glider motion takes place in the visual system, we recorded from the visual cortex of anesthetized macaque monkeys. The results show that a fraction of V1 and V2 neurons were directionally biased for glider motion, and most of these were also directionally biased for standard motion. We also found something puzzling: for individual neurons, the directional bias for three-element glider motion was opposite to the expectation from the psychophysical results. The second investigation focused on the statistics of natural images. Understanding the design of the visual system and how it extracts information requires knowledge of these statistics. As is well-known, second-order statistics of natural images are characterized by a power spectrum of 1/f 2 . In contrast, higher-order statistics are difficult to study because of their high dimensionality. Two-dimensional Hermite (TDH) functions have mathematical properties that recommend them for analysis of image statistics. We applied TDH functions as filters to a set of natural images, and then calculated the statistics of the filter outputs. The analysis yielded a compact and comprehensive description of high-order statistics of natural images, and this description reflects both their phase and amplitude structure. Finally, we discuss how these findings are related to their "causes" in the real world, and how the visual system can take advantage of them.
Author: Qin Hu Publisher: ISBN: Category : Languages : en Pages : 202
Book Description
This thesis explores the relationship of image statistics to visual processing by considering two important aspects: motion processing and static images. Detection of movement is one of the most fundamental tasks performed by our visual system. The essence of motion is spatiotemporal correlation, but how these correlations are processed biologically is not yet fully known. To probe the computations underlying motion perception, we created a new class of motion stimuli, "glider motion," characterized by their third- and fourth-order spatiotemporal correlations. The direction of glider motion cannot be detected by current models for the neural computation of spatiotemporal correlations. Nevertheless, glider stimuli reliably produced a directional motion percept in humans. This implies that the current models for the computations underlying motion processing require modification, as we discuss. To determine where the calculation of glider motion takes place in the visual system, we recorded from the visual cortex of anesthetized macaque monkeys. The results show that a fraction of V1 and V2 neurons were directionally biased for glider motion, and most of these were also directionally biased for standard motion. We also found something puzzling: for individual neurons, the directional bias for three-element glider motion was opposite to the expectation from the psychophysical results. The second investigation focused on the statistics of natural images. Understanding the design of the visual system and how it extracts information requires knowledge of these statistics. As is well-known, second-order statistics of natural images are characterized by a power spectrum of 1/f 2 . In contrast, higher-order statistics are difficult to study because of their high dimensionality. Two-dimensional Hermite (TDH) functions have mathematical properties that recommend them for analysis of image statistics. We applied TDH functions as filters to a set of natural images, and then calculated the statistics of the filter outputs. The analysis yielded a compact and comprehensive description of high-order statistics of natural images, and this description reflects both their phase and amplitude structure. Finally, we discuss how these findings are related to their "causes" in the real world, and how the visual system can take advantage of them.
Author: Sepide Movaghati Publisher: ISBN: Category : Languages : en Pages :
Book Description
"According to classical models of visual information processing, the visual cortex acts as a hierarchical structure in which information flow is unidirectional and transmitted from the primary visual cortex (V1) to higher visual areas. This model considers V1 as an inflexible processing unit, which filters local basic features of visual input, such as the spatial frequency and orientation of the changes in luminance. Contrary to this classical model, recent studies have reported variant activity in V1 based on the presence of higher-order features within visual inputs. In other words, specific characteristics, such as phase content or spatial coherence, potentially modulate the activity of V1. However, the explicit characteristics and the precise timing of their modulatory effect remain unclear. In this study, MagnetoEncephaloGraphy (MEG) was used to test whether the spatiotemporal response to natural images of animals is different from the response to their phase-shuffled counterparts. The frequency content and the total contrast of the images were maintained throughout the phase-scrambling approach. Meanwhile, higher-order image statistics, such as Kurtosis, were manipulated. We investigated the effect of this manipulation on the evoked response of V1 by taking advantage of (1) the high temporal resolution of the MEG technique and (2) the Minimum norm source localization method. The results of this thesis show that the modulation of activity associated with the presentation of natural images appears as early as 50ms post-stimulation. This effect can be observed in the power of the signals recorded by MEG sensors as well as in sources localized in the primary visual cortex. The phase scrambling of visual input increased the amplitude of the early response (30-70ms post-stimulation) and the M170 component (140-200ms post-stimulation). The weaker population response to natural images during the early response is consistent with a model proposing sparse neuronal coding for natural images. The increased M170 activation of V1 to non-natural phase-scrambled images is consistent with the predictive coding model. According to this model, the prediction error in lower order sensory areas increases with decreasing predictability of the sensory input. An increase in the power of gamma frequency (60-80Hz) was observed in response to phase-scrambled images. These findings are in accordance with a previous study that found prediction errors in auditory cortex to be predominantly conveyed via the high (gamma) frequency range. In addition, spatially coherent features in visual input induce a stronger surround suppression effect and reduce the population activity in V1. Therefore, the mechanisms underlying the late modulatory effect on the M170 component may also involve surround suppression in the visual cortex. Our findings suggest that feedback modulates the neuronal response of the primary visual cortex. Further, they identify V1 as a dynamic filter that can be modulated by the higher-order characteristics of visual input, as opposed to previous models that deem V1 to be an inflexible low-level processing unit." --
Author: Bernd Jähne Publisher: Springer Science & Business Media ISBN: 9783540574187 Category : Technology & Engineering Languages : en Pages : 228
Book Description
Image sequence processing is becoming a tremendous tool to analyze spatio-temporal data in all areas of natural science. It is the key to studythe dynamics of of complex scientific phenomena. Methods from computer science and the field of application are merged establishing new interdisciplinary research areas. This monograph emerged from scientific applications and thus is an example for such an interdisciplinaryapproach. It is addressed both to computer scientists and to researchers from other fields who are applying methods of computer vision. The results presented are mostly from environmental physics (oceanography) but they will be illuminating and helpful for researchers applying similar methods in other areas.
Author: Aapo Hyvärinen Publisher: Springer Science & Business Media ISBN: 1848824912 Category : Medical Languages : en Pages : 450
Book Description
Aims and Scope This book is both an introductory textbook and a research monograph on modeling the statistical structure of natural images. In very simple terms, “natural images” are photographs of the typical environment where we live. In this book, their statistical structure is described using a number of statistical models whose parameters are estimated from image samples. Our main motivation for exploring natural image statistics is computational m- eling of biological visual systems. A theoretical framework which is gaining more and more support considers the properties of the visual system to be re?ections of the statistical structure of natural images because of evolutionary adaptation processes. Another motivation for natural image statistics research is in computer science and engineering, where it helps in development of better image processing and computer vision methods. While research on natural image statistics has been growing rapidly since the mid-1990s, no attempt has been made to cover the ?eld in a single book, providing a uni?ed view of the different models and approaches. This book attempts to do just that. Furthermore, our aim is to provide an accessible introduction to the ?eld for students in related disciplines.
Author: Gregory R. Bock Publisher: John Wiley & Sons ISBN: 0470514620 Category : Psychology Languages : en Pages : 256
Book Description
Foremost neurophysiologists and psychophysicists provide pertinent information on the nature of representation at the earliest stages as this will constrain the disposition of all subsequent processing. This processing is discussed in several different types of visual perception.
Author: Rastislav Lukac Publisher: CRC Press ISBN: 1420054538 Category : Technology & Engineering Languages : en Pages : 600
Book Description
A Decade of Extraordinary Growth The past decade has brought a surge of growth in the technologies for digital color imaging, multidimensional signal processing, and visual scene analysis. These advances have been crucial to developing new camera-driven applications and commercial products in digital photography. Single-Sensor Imaging: Methods and Applications for Digital Cameras embraces this extraordinary progress, comprehensively covering state-of-the-art systems, processing techniques, and emerging applications. Experts Address Challenges and Trends Single-Sensor Imaging: Methods and Applications for Digital Cameras presents leading experts elucidating their own accomplishments in developing the technologies reshaping this field. The editor invited renowned authorities to address specific research challenges and recent trends in their particular areas of expertise. The book discusses single-sensor digital color imaging fundamentals, including reusable embedded software platform, digital camera image processing chain, optical filter and color filter array designs. It also details the latest techniques and approaches in contemporary and traditional digital camera color image processing and analysis for various sophisticated applications, including: Demosaicking and color restoration White balancing and color transfer Color and exposure correction Image denoising and color enhancement Image compression and storage formats Red-eye detection and removal Image resizing Video-demosaicking and superresolution imaging Image and video stabilization A Solid Foundation of Knowledge to Solve Problems Single-Sensor Imaging: Methods and Applications for Digital Cameras builds a strong fundamental understanding of theory and methods for solving many of today’s most interesting and challenging problems in digital color image and video acquisition, analysis, processing, and storage. A broad survey of the existing solutions and relevant literature makes this book a valuable resource both for researchers and those applying rapidly evolving digital camera technologies.
Author: Phillip A. Laplante Publisher: CRC Press ISBN: 1351032739 Category : Technology & Engineering Languages : en Pages : 856
Book Description
The Encyclopedia of Image Processing presents a vast collection of well-written articles covering image processing fundamentals (e.g. color theory, fuzzy sets, cryptography) and applications (e.g. geographic information systems, traffic analysis, forgery detection). Image processing advances have enabled many applications in healthcare, avionics, robotics, natural resource discovery, and defense, which makes this text a key asset for both academic and industrial libraries and applied scientists and engineers working in any field that utilizes image processing. Written by experts from both academia and industry, it is structured using the ACM Computing Classification System (CCS) first published in 1988, but most recently updated in 2012.
Author: Tania Pouli Publisher: CRC Press ISBN: 1568817258 Category : Computers Languages : en Pages : 374
Book Description
To achieve the complex task of interpreting what we see, our brains rely on statistical regularities and patterns in visual data. Knowledge of these regularities can also be considerably useful in visual computing disciplines, such as computer vision, computer graphics, and image processing. The field of natural image statistics studies the regularities to exploit their potential and better understand human vision. With numerous color figures throughout, Image Statistics in Visual Computing covers all aspects of natural image statistics, from data collection to analysis to applications in computer graphics, computational photography, image processing, and art. The authors keep the material accessible, providing mathematical definitions where appropriate to help readers understand the transforms that highlight statistical regularities present in images. The book also describes patterns that arise once the images are transformed and gives examples of applications that have successfully used statistical regularities. Numerous references enable readers to easily look up more information about a specific concept or application. A supporting website also offers additional information, including descriptions of various image databases suitable for statistics. Collecting state-of-the-art, interdisciplinary knowledge in one source, this book explores the relation of natural image statistics to human vision and shows how natural image statistics can be applied to visual computing. It encourages readers in both academic and industrial settings to develop novel insights and applications in all disciplines that relate to visual computing.
Author: Narciso Garcia Publisher: Springer ISBN: 3540397981 Category : Computers Languages : en Pages : 359
Book Description
The purpose of VLBV 2003 was to provide an international forum for the d- cussion of the state of the art of visual content processing techniques, standards, and applications covering areas such as: video/image analysis, representation and coding, communications and delivery, consumption, synthesis, protection, and adaptation. The topics of special interest include all the areas relevant to image communications nowadays, from representation and coding to content classi?cation, adaptation, and personalization. A meeting covering such a wide range of topics takes many years to develop. So, please follow a brief story of the evolution of this relevant and specialized forum and of its adaptation to the prevailing interests along time. At the beginning of 1993, the idea of a specialized workshop to discuss topics in advanced image communications came in Lausanne, Switzerland, at a m- ting of the steering committee of the International Picture Coding Symposium. Therefore, the so-called International Workshop on Coding Techniques for Very Low Bit-rate Video VLBV was born as low bit-rate research was considered to be the leading edge. The ?rst workshop was held at the University of Illinois at Urbana-Champaign, USA, in 1993; the second at the University of Essex in Colchester, UK, in April 1994; the third at NTT in Tokyo, Japan, in November 1995;thefourthattheUniversityofLink ̈ oping,Sweden,inJuly1997;the?fthin Urbana (again) in October 1998. Until this last workshop, VLBV life was closely tied with MPEG-4, that is to low bit-rate research.
Author: Qin Hu
Author: Qin Hu
Author: Sepide Movaghati
Author: Bernd Jähne
Author: Aapo Hyvärinen
Author: 
Author: Gregory R. Bock
Author: Rastislav Lukac
Author: Phillip A. Laplante
Author: Tania Pouli
Author: Narciso Garcia