Parallel Processing Of Visual Information Research Articles

Global salient object detection based on multiple visual features

Salient object detection has become an important tool in the fields of computer vision and image processing. In this paper, we propose a novel global salient object detection model based on multiple visual features. Firstly, following the mechanism of parallel processing of visual information, we obtain three saliency maps based on colour, intensity difference, and spatial distribution. Secondly, introducing the concept of spatial gaze point of an image, we calculate three Gaussian weighted maps based on three saliency maps to constrain their background noise and get three weighted saliency maps. Finally, the final saliency map is produced by fusing three weighted saliency maps. We compare our method with 10 state-of-the-art saliency methods on a public dataset, and the results show that our method outperforms the other 10 methods.

The effect of discomfort glare on parallel processing of visual information

AbstractPurpose When a person is confronted with a bright light source against a dim background, scattered light can cause visual discomfort in addition to reducing object contrast on the retina. Previous work addressing the effect of glare on visual performance has focused mainly on the reduction in retinal image contrast (known as disability glare) rather than the often accompanying experience of discomfort glare. This study examines how parallel processing of visual information is affected under conditions of discomfort glare.Methods Light scatter sources surrounded a monitor on which the subject carried out standard contrast acuity (CA) tasks. All CA tasks were carried out with and without discomfort glare (estimated in a prior experiment). In the parallel task, a fovea target (Landolt C) and four peripheral stimuli (three distractors and one target) were presented concurrently; the subject had to indicate both the orientation of the gap for the foveal target and the quadrant containing the peripheral target. The size of the peripheral target was increased relative to the foveal target to ensure that all targets were equally detectable. This was done separately with and without discomfort glare, thus negating any influence of disability glare.Results The presence of discomfort glare degrades visual performance in the parallel processing task even when adjustments are made to cancel out the effects of disability glare.Conclusion Studies that have focused only on disability glare may be underestimating the adverse effect glaring light sources can have on visual performance, particularly in real world scenarios where attention to objects in both the fovea and periphery is often required.

Corpus Callosal Microstructural Integrity Influences Interhemispheric Processing: A Diffusion Tensor Imaging Study

Normal aging and chronic alcoholism result in disruption of brain white matter microstructure that does not typically cause complete lesions but may underlie degradation of functions requiring interhemispheric information transfer. We examined whether the microstructural integrity of the corpus callosum assessed with diffusion tensor imaging (DTI) would relate to interhemispheric processing speed. DTI yields estimates of fractional anisotropy (FA), a measure of orientation and intravoxel coherence of water diffusion usually in white matter fibers, and diffusivity (<D>), a measure of the amount of intracellular and extracellular fluid diffusion. We tested the hypothesis that FA and <D> would be correlated with (i) the crossed-uncrossed difference (CUD), testing visuomotor interhemispheric transfer; and (ii) the redundant targets effect (RTE), testing parallel processing of visual information presented to each cerebral hemisphere. FA was lower and <D> higher in alcoholics than in controls. In controls but not alcoholics, large CUDs correlated with low FA and high <D> in total corpus callosum and regionally in the genu and splenium. In alcoholics but not controls, small RTEs, elicited with equiluminant stimuli, correlated with low FA in genu and splenium and high <D> in the callosal body. The results provide in vivo evidence for disruption of corpus callosum microstructure in normal aging and alcoholism that has functional ramifications for efficiency in interhemispheric processing.

Parallel interhemispheric processing in aging and alcoholism: relation to corpus callosum size

We tested parallel processing of visual information using the redundant targets effect (RTE) in 12 alcoholics and 13 matched controls. The paradigm was a simple reaction time (RT) task with targets presented in the same (uncrossed), opposite (crossed), or both (redundant) visual-fields. In older alcoholics (>50 years) the RT gain invoked by redundant targets did not exceed probability measures, suggesting compromised interhemispheric processing of parallel information in this subgroup compared with controls or younger alcoholics. The difference between crossed and uncrossed reaction times (CUD), an index of interhemispheric transfer time (ITT), was greater in older than younger subjects. Moreover, the CUD was negatively correlated with the corpus callosum (CC) total area and body in controls, supporting the concept of a structure-function relationship of interhemispheric transfer. This relationship was not found in alcoholics, although the midsagittal area of the CC, genu, and body but not intracranial volume (ICV), was significantly smaller in alcoholics than controls. These results suggest that chronic alcohol abuse together with advancing age exert subtle disruption on parallel interhemispheric processing reliant on callosal connections.

High-resolution spatio-temporal mapping of visual pathways using multi-electrode arrays

The parallel processing of visual information was studied with penetrating microelectrode arrays. We studied the high-resolution visuotopic organization of cat primary visual cortex, and the encoding of simple visual stimuli by ensembles of ganglion cells in the isolated turtle retina. The high-resolution visuotopic organization of visual cortex is non-conformal. Regions of visual cortex separated by 400 μ may have receptive field centers that are separated by as much as 3°, or they may superimpose. Ganglion cells are ‘generalists’, and are poor specifiers of the color of full field visual stimuli. Groups of ‘luminosity’ type ganglion cells can assist in the specification of stimulus color, but even individual ‘chromatic’ ganglion cells are not capable of quality color specification. These basic studies have relevance to the development of visual neuroprostheses based upon electrical stimulation of the retina and cortex.

Parallel processing and image analysis in the eyes of mantis shrimps.

The compound eyes of mantis shrimps, a group of tropical marine crustaceans, incorporate principles of serial and parallel processing of visual information that may be applicable to artificial imaging systems. Their eyes include numerous specializations for analysis of the spectral and polarizational properties of light, and include more photoreceptor classes for analysis of ultraviolet light, color, and polarization than occur in any other known visual system. This is possible because receptors in different regions of the eye are anatomically diverse and incorporate unusual structural features, such as spectral filters, not seen in other compound eyes. Unlike eyes of most other animals, eyes of mantis shrimps must move to acquire some types of visual information and to integrate color and polarization with spatial vision. Information leaving the retina appears to be processed into numerous parallel data streams leading into the central nervous system, greatly reducing the analytical requirements at higher levels. Many of these unusual features of mantis shrimp vision may inspire new sensor designs for machine vision.

The persistence of position

I describe a signal coined position persistence that stores information about the last seen position of an object. Position persistence is not the same as visible persistence, although some of its properties are similar. The duration of position persistence is such that objects visible briefly always generate a position signal for at least 180 ms. The signal is not affected by the intensity of the object, nor of the background. Position persistence decreases with increasing speed, but does not depend on retinal eccentricity. Finally, the persisting signal is not tightly bound to the object that causes it. The signal contains no information on the colour of the object, whereas shape information may become represented after approximately 100 ms. The existence of this signal is interpreted as a psychophysical signature of the parallel processing of visual information.

Aging and Feature Search: The Effect of Search Area

The preattentive system involves the rapid parallel processing of visual information in the visual scene so that attention can be directed to mean ingful objects and locations in the environment. This study used the feature search methodology to examine whether there are aging-related deficits in parallel-processing capabilities when older adults are required to visually search a large area of the visual field. Like young subjects, older subjects displayed flat, near-zero slopes for the Reaction Time 2 Set Size function when searching over a broad area (30 radius) of the visual field, implying parallel processing of the visual display. These same older subjects exhibited impairment in another task, also dependent on parallel processing, performed over the same broad field area; this task, called the useful field of view test, has more complex task demands. Results imply that aging-related breakdowns of parallel pro cessing over a large visual field area are not likely to emerge when required responses are simple, there is only one task to perform, and there is no limitation on visual inspection time.

Amplitude Scaling Compensates for Serial Delays in Correcting Eye and Arm Movements.

Spatial and metrical parameters of the eye and arm movements made by human subjects (N = 7) in response to visual targets that were stepped unexpectedly either once (single step) or twice (double step) were studied. For the double-step, the displacement of a visual target was decreased or increased in amplitude at intervals before and during a movement. Provided the second target step occurred more than 100 ms before the onset of movement, the amplitude of the subjects' first response was altered in the direction of the new target location. But this amplitude scaling was not always sufficient to reach the new target location, and a second corrective response was required. The latency in producing this second response was greatly increased above reaction time latencies of movements to single-step targets, especially when the target change occurred 100 ms or more before movement onset. These findings suggest that even though serial processing limitations delay the production of a second corrective response, continuous parallel processing of visual information enables the amplitude of the first response to be altered with minimal delay. This enables some degree of real-time continuous control by the visuomotor control system.

VISION, BRAIN, AND BEHAVIOUR IN BIRDS

Part 1 The avian eye view: introduction, Sally A. McFadden producing the image, Graham R. Martin exploring the image, Hans-Ortwin Nalbach et al constructing the three-dimensional image, Sally A. McFadden the visual capabilities of birds, William Hodos colour vision of birds, Francisco J. Varela et al. Part 2 Functional anatomy of the avian visual system: introduction, Onur Gunturkun the avian visual system and the evolution of the neocortex, Toru Shimizu and Harvey J. Karten anatomy of the avian thalamofugal pathway, Onur Gunturkun et al the organization of the tectofugal pathway in birds - a comparative review, Jurgen Engelage and Hans-Joachim Bischof binocular processing in frontal-eyed birds, G. Casini et al. Part 3 Development of the avian visual system: introduction, Hans-Joachim Bischof developmental anatomy of the chick retinotectal projection, Jorg Mey and Solon Thanos development, plasticity, and differential organization of parallel processing of visual information in birds, G. Fontanesi et al development of the tectofugal visual system of normal and deprived zebra finches, Kathrin Herrmann and Hans-Joachim Bischof developmental mechanisms of lateralization, Lesley J. Rogers and Patrice Adret. Part 4 Visuomotor mechanisms: introduction, H. Philip Zeigler eye movements, head movements, and gaze stabilization in birds, Josh Wallman and Juan-Carlos Letelier sensorimotor mechanisms and pecking in the pigeon, H. Philip Zeigler et al control of pecking response topography by stimulus-reinforcer and response-reinforcer contingencies, Robert W. Allan visual mechanisms of prey capture in water birds, Gadi Katzir. Part 5 Vision and cognition: introduction, Jacky Emmerton lateralization and strategies of viewing in the domestic chick, R.J. Andrew and M. Dharmaretnam two eyes and one world - studies of interocular and intraocular transfer in birds, Monika Remy and Shigeru Watanabe what can we learn from experiments on pigeon concept discrimination?, Shigeru Watanabe et al beyond sensation - visual cognition in pigeons, Jacky Emmerton and Juan D. Delius vision, cognition, and the avian hippocampus, Verner P. Bingman.

Dendritic sampling across processing streams in monkey striate cortex.

Cytochrome oxidase (CO) dense blobs in primate striate cortex provide a striking example of parallel processing of visual information. The level of isolation of the blobs from the surrounding interblob tissue was investigated in the present study by combining CO staining with Golgi impregnation of dendritic arbors in the same tissue sections. The data are based on material from two marmoset and three squirrel monkeys. The analysis was conducted on two types of Golgi preparations. In the first preparation, dense networks of overlapping dendrites were impregnated over blob margins. The results of analyzing these networks with transmission and confocal microscopy revealed that dendritic arbors penetrate freely through blob margins. Statistical analysis revealed that the density of dendritic crossings at blob margins was similar to that found at blob and interblob centers. In the second type of Golgi preparation, single, isolated neurons were impregnated. Studies of such neurons revealed occasional examples of dendritic arbors that appeared to reflect back from blob margins, but counter examples were equally abundant. Bias index analysis indicated that dendritic arbors were generally unaffected by the presence of a nearby blob margin. Scanning a large number of impregnated arbors indicated that at least half of the population of blob-related neurons had dendrites in both blob and interblob territory. Under the conditions of free dendritic penetration of blob margins, the sole factor that determines the level of blob/interblob mixing appears to be the relationship between blob size and the dendritic spread of blob neurons. Interestingly, in both the marmoset and squirrel monkeys this size ratio is similar despite a large difference in their cortical surface area. Thus, it is hypothesized that blob size is optimally matched to the dendritic span so as to create a smooth transition of dendritic sampling from blob to interblob-related processing streams.

Neural mapping and space-variant image processing

The mapping of sensory surfaces onto cortical areas as well as maps in inter-areal projections are classified and discussed in terms of space-variant image processing. The neurobiological phenomena discussed in this context include retinotopic maps, columnar organization, interactions between a retinotopic map and a uniform operation in a cortical area, and patchy or segregated projections of two or more areas to a common target area. We present a mathematical framework for describing the interaction of neural mappings and local image processing operations which allows functional interpretations. In an example from visual navigation, we show that neural maps are powerful tools for the parallel processing of visual information. Since mappings of the various types result in a spatial encoding of arbitrary stimulus information, image processing operations can be applied to information from other modalities or for higher-level problems as well. For this aspect of neural mapping, we adopt the term parametric mapping.