Showing papers on "Human visual system model published in 1982"

A Demonstration of the Visual Importance and Flexibility of Spatial-Frequency Amplitude and Phase

Abstract: To establish how little information the human visual system requires for recognition, common objects were digitally manipulated in the Fourier domain. The results demonstrate that it is not only possible, but also quite efficient, for a (biological) visual system to exist with very few phase relationships among the component spatial frequencies of the (retinal) image. A visual example is then presented which illustrates how certain phase relationships can hinder, or completely eliminate, the recognition of visual scenes.

A model of the early stages of the human visual system: Functional and topological transformations performed in the peripheral visual field

Abstract: A model of the early stages of the visual system is presented, with particular reference to the region of the visual field outside the fovea and to the class of retinal and lateral geniculate nucleus cells which are most active in the processing of pattern information (X-cells). The main neuroanatomical and neurophysiological properties taken into account are: the linear increase of the receptive fields diameter with eccentricity, the constancy of the overlap factor and the topological transformation operated upon the retinal image by the retino-cortical connection. The type of filtering taking place between the retina and the visual cortex is analyzed and some simulations are presented. It is shown that such a filtering is of a band-pass space variant type, with center frequencies that decrease from the center (i.e. the fovea) toward the periphery of the visual field. This processing is "form invariant" under linear scaling of the input. Moreover, considering the properties of the retino-cortical connection, it is shown that the "cortical images" undergo simple shifts whenever the retinal images are scaled or rotated.

Mechanisms of interpolation in human spatial vision.

Abstract: Recently, interest has revived in the classical problems of vernier and stereoscopic acuity. The precision with which observers can align two bars in a vernier task is as high as 2 sec arc, which is much finer than the grain of the retinal receptor mosaic, where the cones are separated by 20–30 sec arc even at their most densely packed. Thus, it seems that the visual system may be able to interpolate between the discrete samples provided by the retinal mosaic; recently, several possible mechanisms have been proposed1,2. Mathematically there is no mystery about interpolation. The optical system of the eye acts as a low-pass filter removing all frequencies above ∼60 cycles deg−1, so it follows from the sampling theorem that the retinal image can be fully represented by sampling at a frequency of 120 cycles deg−1, which is approximately that of the receptor mosaic. In principle, therefore, discrete sampling by the retinal mosaic does not remove information from the image. Indeed, the low-pass filtering action of the eye implies that acuity would not be lost, even if the signal itself were divided into discrete samples, before observation. We report here that human observers can locate the spatial position of a periodic visual pattern with a precision as high as 5–10 sec arc, even though the pattern is coarsely sampled at an interval over 10 times that amount. This suggests that the human visual system can construct a surprisingly accurate representation of a pattern from discrete samples, despite the samples being sufficiently widely spaced to be visually resolved.

Multiplication noise in the human visual system at threshold

Abstract: Several kinds of light used in vision experiments produce photon statistics that are distinctly non-Poisson. Representative examples are light from a cathode-ray tube and an image-intensifier device. For the class of vision experiments in which the photon statistics play an important role, excess fluctuations produced by such light sources can alter the observed results and obscure the visual mechanisms being studied. They must therefore be accounted for in a proper way. We use the results of a Hecht-Shlaer-Pirenne type experiment, carried out with modulated Poisson light, to illustrate the point. Sensitivity and modulation depth, as well as sensitivity and reliability, are shown to be traded against each other. Finally, we demonstrate that number-state light, which is comprised of photons of an ideal kind, provides the ultimate tool for extracting information about the intrinsic noise distribution in the visual system at threshold. The state of the art in producing such light is discussed.

A contour-texture approach to picture coding

Abstract: A new image coding technique, based on a simplified description of a region-composed image is presented. The regions are extracted from the original image by an adaptive psychovisually oriented segmentation process. The picture elements (pels) inside the regions represent the texture information of the image whereas the region boundaries points represent the contour information. The data-adaptive character of this method as well as the use of some psychovisual characteristics of the human visual system lead to promising results in data compression.

Towards A Complete Model Of Photopic Visual Threshold Performance

Abstract: Based on a wide variety of fragmentary evidence taken from psycho-physics, neurophysiology and electron microscopy, it has been possible to put together a very widely applicable conceptual model of photopic visual threshold performance. Such a model is so complex that a single comprehensive mathematical version is excessively cumbersome. It is, however, possible to set up a suite of related mathematical models, each of limited application but strictly known envelope of usage. Such models may be used for assessment of a variety of facets of visual performance when using display imagery, including effects and interactions of image quality, random and discrete display noise, viewing distance, image motion, etc., both for foveal interrogation tasks and for visual search tasks. The specific model may be selected from the suite according to the assessment task in hand. The paper discusses in some depth the major facets of preperceptual visual processing and their interaction with instrumental image quality and noise. It then highlights the statistical nature of visual performance before going on to consider a number of specific mathematical models of partial visual function. Where appropriate, these are compared with widely popular empirical models of visual function.

Modelling perceptual distortion: amplitude and phase transmission in the human visual system.

Abstract: In this paper we consider the complementary properties of image amplitude and phase components and their role in normal and amblyopic vision. Specific two-dimensional filtered images are included to demonstrate the possible perceptual distortions due to both amplitude and phase disturbances. Finally, the relationships between amplitude and phase filters are discussed with reference to the types of receptive field distributions expected to underly the normal and abnormal cases.

SSAM: Solid-State Automated Microscope

Abstract: A new microscope system that is designed to provide images for a computer has been built and tested. This system differs from previous computerized microscopes in that the fundamental design parameters have been tuned to the computer as the receiver of the image instead of the human visual system. This solid-state automated microscope system (SSAM) simultaneously provides wide-field (2mm), high-resolution (0.5 ?), high signal-to-noise images (53 dB) at data rates of 5 X 106 pixels/s. Various methods have been developed and used to test the design specifications of the system against the actual performance.

Theory and Model of the Human Global Analysis of Visual Structure

Abstract: An introduction is given to a theory of global structure analysis in early visual information processing. The theory relates retinal, visual cortical, retino-collicular, cortico-collicular, and oculomotor mechanisms to specific properties of the (human) visual structure analysis. Information theoretic modeling of the retina, particularly rate distortion theoretic interpretations of the retina-oculomotor system functions reveal both behavior and information processing task optimality. From a structure system point of view these properties bring forth a global structure identification function and its supporting global data systems which extract information from the object system. A general formulation of the structure identification function and its data systems is given in terms of partial correlations of perceptual similarity. The theory has been implemented for dot patterns through a set of computer programs. This computer model determines structure?perceptually important global features, their "strength," and the pattern regions that give rise to these features. Psychophysical measurements were necessary to set the "human" parameters of the model. Global analysis of visual structure is considered to be essentially a structure segmentation and feature classification task that precedes detailed information extraction. Theory-based interpretation of the functional organization of the sensory, motor, and neurophysiological components of the visual system leads to expectations concerning a scan strategy that directs the retina towards regions of high structure information density.

Reconstruction of spatial information in the human visual system.

Abstract: The remarkable capacity of an observer to perceive and recognize objects and forms independent of the exact nature of their components can be described as a process of reconstruction. We have measured the ability of human observers to recognize square-wave or sinusoidal gratings when presented as a pattern of regularly taken spatial samples (see Fig. 1). To recognize the waveforms, the observer must visually reconstruct them from the samples—a process which can be described by the Shannon–Whittaker theorem1,2 of sampling and which Barlow suggests may be carried out by the numerous stellate cells of the visual cortex3. We report here that at low spatial frequencies, relatively more sample lines per spatial cycle were needed for wave recognition than were needed at higher frequencies. However, a square-wave grating was still recognized easily even when it was sampled at a rate at which its third harmonic could not be recognized when presented alone. At high spatial frequencies the square wave was identified when the sampling rate was so low that it caused the third harmonic to be under-sampled. This contradicts the idea that a complex wave is analysed by parallel spatial frequency channels4,5 and emphasizes the capacity of the visual system to use signal features other than the harmonic frequency components of an image to recognize it.

Artificial Intelligence Approaches to Image Understanding

Abstract: The past decade has seen considerable progress in the development of computer vision within Artificial Intelligence. Attention has shifted from restrictions on the domain of application of a vision system to restrictions on the visual abilities studied. Mathematical analyses have been offered for some of the elements of visual perception, such as the relationship between image irradiance and scene radiance, the location of important intensity changes and motion primitives. In each case, it is observed that the information in the image only partially constrains the interpretation of the image, and further constraints are sought. The constraints embody commitments about the way the world is, at least most of the time. For example, the world mostly consists of smooth surfaces, and scenes are mostly viewed from general position, free of accidental alignments. Perceptual abilities such as stereopsis, lightness determination, shape from shading and from texture, require that the appropriate constraints be uncovered and appropriately expressed. Representations have been developed that make explicit the information computed by a perceptual ability. Examples include the Primal Sketch, the Reflectance Map, and object representations based on generalised cones. The isolation of representations has lead to a view of visual perception as the process of constructing instances of a sequence of representations. The input to a particular process is often not the image per se, but a representation of the information computed by a number of processes. It is this observation which most strongly distinguishes image understanding from conventional pattern recognition. A number of sample image understanding systems are described, including edge detection, shape from shading, binocular and photometric stereo, optical flow, directional selectivity, surface reconstruction through interpolation and the representation of objects by primitive volumes. In some cases, it has been possible to directly relate the theory embodied in the program to animate visual systems. In some cases it has been possible to develop important practical applications for example, industrial inspection and bin picking in robotics, and monitoring airfields or terrain for changes in usage. Finally, in some cases it has been possible to construct hardware realizations of theories to achieve real time performance.

Chapter 10 – Segmentation

Abstract: Publisher Summary In image compression or enhancement, the desired output is a picture— an approximation to, or an improved version of, the input picture. Another major branch of picture-processing program deals with image analysis or scene analysis. This chapter discusses picture and scene segmentation techniques. Some segmentation operations can be applied directly to any picture; others can only be applied to a picture that has already been partially segmented as they depend on the geometry of the parts that have already been extracted from the picture. For example, a chromosome picture can be (crudely) segmented by thresholding its gray level. Once this has been done, further segmentation into individual chromosomes can be attempted, based on connectedness, size, and shape criteria. There is no single standard approach to segmentation. The perceptual processes involved in segmentation of a scene by the human visual system, such as the Gestalt laws of organization, are not yet well understood.

Human pattern detection and recognition in the search for extraterrestrial intelligence

Abstract: Potential radio signals from extraterrestrial civilizations will be weak with respect to background noise. We investigated people’s ability (N = 18) to visually detect and recognize signals under noise conditions that simulate those expected to be present in an actual search. Linear signals were detected and identified better than pulses and wavy lines. In addition, observers sometimes agreed on the location of pseudosignals within a visual noise display (no actual signal present). Implications are noted for the search process as well as for advancing knowledge of the human visual system’s capacity to detect and recognize signals.

Applications of the Nonlinear, Visual Operator “Lateral Inhibition” in Image Processing

Abstract: There have been some attempts to emulate a possible focussing mechanism of the human eye in technical optics, simulating lateral inhibition [1]. To study the basics of practical applications, this physiological principle was optically and information theoretically investigated in the human visual system (HVS).

Chapter XVI Quantitative Objective Study of Human Visual Perception and Recognition

Abstract: Publisher Summary This chapter discusses the quantitative objective study of human visual perception and recognition. The human visual system represents a magnificent tool of pattern recognition and information processing ,including the level of conscious cognitive phenomena. However, quantitative and computer approaches open new research possibilities, mainly in (a) stimulus generation and control, (b) detection of meaningful signals and data processing, and (c) modeling and formulation of acceptable explanative hypothesis. A review of three types of experiments dealing with visual perception and cognition in humans from the objective and qualitative points of view is presented. Experiments using reversible figures as stimuli demonstrated some specific aspects of geometrical feature perception and pointed to the general problematic of real object's representation in geometrical terms. By means of computerized bioelectrical signal detection and classification procedures, a physiological correlate of some human cognitive processes is described. Scanning eye movement control is analyzed by means of recording and computing techniques, and the significance of both local geometrical features and central semantic processes delineated in human pattern recognition and visual perception.

Extended Graphic Functions of the A-IDAS System for Visual Design

Abstract: Recently a remarkable progress was seen in computer graphics. In addition to a traditional vector-type graphic device, the debut and popularization of a raster-type color one signify the said progress. Computer graphics are expanding not only in CAD fields but also in visual application fields as landscape simulation, computer art, animation, etc. With this background, we introduce a visual design system which applys computer graphics to a visual environment planning/designing area.

Pilot Task Profiles, Human Factors, And Image Realism

Abstract: Computer Image Generation (CIG) visual systems provide real time scenes for state-of-the-art flight training simulators. The visual system reauires a greater understanding of training tasks, human factors, and the concept of image realism to produce an effective and efficient training scene than is required by other types of visual systems. Image realism must be defined in terms of pilot visual information reauirements. Human factors analysis of training and perception is necessary to determine the pilot's information requirements. System analysis then determines how the CIG and display device can best provide essential information to the pilot. This analysis procedure ensures optimum training effectiveness and system performance.