ITC/USA 2010 Conference Proceedings / The Forty-Sixth Annual International Telemetering Conference and Technical Exhibition / October 25-28, 2010 / Town and Country Resort & Convention Center, San Diego, California / Aerial image collections have experienced exponential growth in size in recent years. These high resolution images are often viewed at a variety of scales. When an image is displayed at reduced scale, maximum quantization step sizes for visually lossless quality become larger. However, previous visually lossless coding algorithms quantize the image with a single set of quantization step sizes, optimized for display at the full resolution level. This implies that if the image is rendered at reduced resolution, there are significant amounts of extraneous information in the codestream. Thus, in this paper, we propose a method which effectively incorporates multiple quantization step sizes, for various display resolutions, into the JPEG2000 framework. If images are browsed from a remote location, this method can significantly reduce bandwidth usage by only transmitting the portion of the codestream required for visually lossless reconstruction at the desired resolution. Experimental results for high resolution color aerial images are presented.

This research explores the differences between how game engine cameras and the human visual system (HVS) render colour. The study is motivated by a two part research question: will HVS colours or game camera colours be preferred when experiencing a virtual environment from a 1st-person perspective and how does light intensity relate to preference? While previous research defines perceptual processes which influence the interpretation of colour information this study advances the understanding of how these theories may be applied to 3D colour grading.When evaluating the two colour modes with a combination of quantitative data and qualitative reflections it was possible to establish a correlation between preference and light intensity, in the sense that HVS colours were preferred in high illumination and camera colours in low. The findings implicate that in order to be well received the colours of a virtual environment need to be adjusted according to illumination.

Background: Sit-to-stand is a challenging task as it requires the transition from a large 3-point base of support to a small 2-point base of support while simultaneously controlling anteroposterior and vertical body acceleration. Age-related morphological changes in both the visual and vestibular system could impair the ability to extract and interpret sensory information necessary for motor control in older adults, which can increase instability and the risk of falls. The purpose of this study is to understand the effects of aging on visual-vestibular interaction for maintaining stability during sit-to-stand. Methods: Fifteen younger (age=22.5±1.1) and fifteen older (age=73.9±5.3) healthy adults were asked to stand from a sitting position as quickly as possible. Vestibular input was manipulated using percutaneous bipolar galvanic vestibular stimulation where threshold intensity was individually calculated for each participant during quiet stance with eyes closed. Galvanic vestibular stimulation was applied at both threshold (1xGVS) and 2-times the participant’s threshold intensity (2xGVS). Visual conditions included eyes opened, wearing custom-made vision blurring goggles, or eyes closed. Outcome measures included a global measure of performance (transition phase duration), mediolateral stability (peak-to-peak trunk roll angle, mediolateral center of mass displacement, mediolateral center of pressure displacement) and anteroposterior stability (peak braking force, peak-to-peak trunk pitch angle, and peak anteroposterior center of mass velocity). Results: When vision was suboptimal (blurring goggles), older adults had significantly longer transition phase duration than younger adults (p<0.05). Older adults demonstrated greater mediolateral instability than younger adults. When vision was absent, trunk roll angle was significantly greater with 1xGVS than 2xGVS (p<0.05). Mediolateral center of mass displacement was greater when vision was absent than when vision was available, irrespective of age (p<0.05). No effects of age, vision or galvanic vestibular stimulation were seen in peak braking force, trunk pitch angle, and peak anteroposterior center of mass velocity (p>0.05). Conclusion: Regardless of age, visual inputs were more critical to maintain stability during sit-to-stand than vestibular inputs. Differences between younger adults and older adults were only seen in the mediolateral direction. Despite having greater mediolateral instability, older adults utilized similar strategies as younger adults to overcome sensory perturbations during sit-to-stand. / Thesis (Master, Rehabilitation Science) -- Queen's University, 2013-09-20 23:22:16.695

<p>A halftoning method with Neugebauer Primary Area Coverage direct binary search (NPAC-DBS) is developed. With the optimized human visual system (HVS) model, we are able obtain homogeneous and smooth halftone colored image. The halftoning is based on separating the colored image represented in Neugebauer Primary in three channels based on human visual system, with swap-only DBS, we arrange the dots to bring the error metric to its minimum and the optimized halftone image is obtained. The separation of chrominance HVS filters between red-green and blue-yellow channels allows us to represent HVS more accurately. Color halftone images generated using this method and method of using traditional screening methods are compared.</p> <p>In order to speed up the halftone process with similar quality of NPAC-DBS, we developed PARAWACS screens for color halftoning. PARAWACS screen is designed level by level using DBS. With PARAWACS screen, we can create halftone using simple pixel by pixel comparison with the merit of DBS. We further optimized the screen to achieve the best quality.</p> <p>Next, a novel halftoning method that we call Ink-Saving, Single-Frequency, Single-Angle, Multi-Drop (IS-SF-SA-MD) halftoning is introduced. The application target for our algorithm is high-volume production ink-jet printing in which the user will value a reduction in ink usage. Unlike commercial offset printing in which four-colorant printing is achieved by rotating a single screen to four different angles, our method uses a single frequency screen at a single angle, and it depends on accurate registration between colorant planes to minimize dot-overlap especially between the black (K) colorant and the other colorants (C, M, and Y). To increase the number of gray levels for each colorant, we exploit the multidrop capabilities of the target writing system. We also use the hybrid screening method to yield improved halftone texture in the highlights and shadows. The proposed method can help preserve ink significantly.</p>

Abstract This thesis describes research into the field of image restoration. Restoration is a process by which an image suffering some form of distortion or degradation can be recovered to its original form. Two primary concepts within this field have been investigated. The first concept is the use of a Hopfield neural network to implement the constrained least square error method of image restoration. In this thesis, the author reviews previous neural network restoration algorithms in the literature and builds on these algorithms to develop a new faster version of the Hopfield neural network algorithm for image restoration. The versatility of the neural network approach is then extended by the author to deal with the cases of spatially variant distortion and adaptive regularisation. It is found that using the Hopfield-based neural network approach, an image suffering spatially variant degradation can be accurately restored without a substantial penalty in restoration time. In addition, the adaptive regularisation restoration technique presented in this thesis is shown to produce superior results when compared to non-adaptive techniques and is particularly effective when applied to the difficult, yet important, problem of semi-blind deconvolution. The second concept investigated in this thesis, is the difficult problem of incorporating concepts involved in human visual perception into image restoration techniques. In this thesis, the author develops a novel image error measure which compares two images based on the differences between local regional statistics rather than pixel level differences. This measure more closely corresponds to the way humans perceive the differences between two images. Two restoration algorithms are developed by the author based on versions of the novel image error measure. It is shown that the algorithms which utilise this error measure have improved performance and produce visually more pleasing images in the cases of colour and grayscale images under high noise conditions. Most importantly, the perception based algorithms are shown to be extremely tolerant of faults in the restoration algorithm and hence are very robust. A number of experiments have been performed to demonstrate the performance of the various algorithms presented.

Most computational models of neurons assume that their electrical characteristics are of paramount importance. However, all long-term changes in synaptic efficacy, as well as many short-term effects, are mediated by chemical mechanisms. This technical report explores the interaction between electrical and chemical mechanisms in neural learning and development. Two neural systems that exemplify this interaction are described and modelled. The first is the mechanisms underlying habituation, sensitization, and associative learning in the gill withdrawal reflex circuit in Aplysia, a marine snail. The second is the formation of retinotopic projections in the early visual pathway during embryonic development.

ITC/USA 2013 Conference Proceedings / The Forty-Ninth Annual International Telemetering Conference and Technical Exhibition / October 21-24, 2013 / Bally's Hotel & Convention Center, Las Vegas, NV / This paper described the details of subjective validation for visually lossless compression of stereoscopic 3 dimensional (3D) images. The subjective testing method employed in this work is adapted from methods used previously for visually lossless compression of 2 dimensional (2D) images. Confidence intervals on the correct response rate obtained from the subjective validation of compressed stereo pairs provide reliable evidence to indicate that the compressed stereo pairs are visually lossless.

ITC/USA 2012 Conference Proceedings / The Forty-Eighth Annual International Telemetering Conference and Technical Exhibition / October 22-25, 2012 / Town and Country Resort & Convention Center, San Diego, California / This paper proposes a method of measuring visibility thresholds for quantization distortion in JPEG2000 for compression of stereoscopic 3D images. The crosstalk effect is carefully considered to ensure that quantization errors in each channel of stereoscopic images are imperceptible to both eyes. A model for visibility thresholds is developed to reduce the daunting number of measurements required for subjective experiments.

Cette thèse vise à proposer une nouvelle architecture de codage vidéo qui s’inspire du système visuel des mammifères et de la rétine. La rétine peut être considérée comme une machine intelligente qui traite le stimulus visuel de façon très efficace. De ce fait, elle représente une grande source d'inspiration pour développer de nouveaux systèmes de traitement d’image. Il y a plusieurs raisons pour cela : elle consomme peu d’énergie, elle traite des entrées haute résolution et sa façon de transformer et d’encoder de manière dynamique le stimulus visuel dépasse les normes actuelles. Nous avons souhaité étudier et proposer un codec vidéo inspiré de la rétine. L’algorithme proposé a été appliqué à un flux vidéo d'une manière simple, suivant le principe des standards de codage MJPEG ou MJPEG2000. Cette approche permet au lecteur d’étudier et d’explorer tous les avantages du traitement dynamique de la rétine en termes de compression et de traitement d’image. La performance actuelle du codec que nous avons développé est très prometteuse. Les résultats montrent des performances supérieures à MJPEG pour des débits inférieurs à 100 kbps et MPEG-2 pour des débits supérieurs à 70 kpbs. De plus, à faibles débits, le codec proposé décrit mieux le contenu de la scène d’entrée. De nombreuses perspectives sont proposées afin d'améliorer ce codec inspiré de la rétine qui semblent conduire à un nouveau paradigme en compression vidéo. / The goal of this thesis is to propose a novel video coding architecture which is inspired by the mammalian visual system and the retina. If one sees the retina as a machine which processes the visual stimulus, it seems an intelligent and very efficient model to mimic. There are several reasons to claim that, first of all because it consumes low power, it also deals with high resolution inputs and the dynamic way it transforms and encodes the visual stimulus is beyond the current standards. We were motivated to study and release a retina-inspired video codec. The proposed algorithm was applied to a video stream in a very simple way according to the coding standards like MJPEG or MJPEG2000. However, this way allows the reader to study and explore all the advantages of the retina dynamic processing way in terms of compression and image processing. The current performance of the retina-inspired codec is very promising according to some final results which outperform MJPEG for bitrates lo wer than 100 kbps and MPEG-2 for bitrates higher than 70 kpbs. In addition, for lower bitrates, the retina-inspired codec outlines better the content of the input scene. There are many perspectives which concern the improvement of the retina-inspired video codec which seem to lead to a groundbreaking compression architecture. Hopefully, this manuscript will be a useful tool for all the researchers who would like to study further than the perceptual capability of the mammalian visual system and understand how the structure and the functions of the retina can in practice improve the coding algorithms.

Retinal terminal profiles in the goldfish optic tectum were identified electron microscopically after (1) labeling with horseradish peroxidase and (2) in the early stages of degeneration in short‐term eye enucleates. All labeled terminals shared certain common morphological characteristics which were identical to those of a population of terminals in normal tecta. Terminals of this type disappeared 30 days after enucleation of the contralateral eye. Retinal terminal presynaptic profiles were characterized by (1) round and oval synaptic vesicles; (2) mitochondria with irregular, randomly oriented cristae, large intracristal spaces, dilated membrane spaces, and primarily light matrices; (3) a wide range in profile area, 0.06–6.82 μm2; (4) large numbers of synaptic vesicles per profile area 168± 33 synaptic vesicles per μm2; (5) asymmetric synapses; and (6) multiple synaptic contacts (1.46 ± 0.73 per terminal profile). The postsynaptic elements included both dendritic and, less commonly, pleomorphic vesicle‐containing profiles. The majority of postsynaptic dendritic profiles were small (0.01–0.40 μm2). Serial synaptic contacts were occasionally seen. The combination of vesicular and mitochondrial morphology (1 and 2 above) was necessary and sufficient to establish the retinal origin of a terminal, but use of such criteria would underestimate the number of retinotectal terminals by omitting those which did not have a mitochondrion in the plane of section. The number of such terminals was calculated from independent measurements, and the total number of retinal terminal profiles per area of neuropil was estimated.