Thesis (M.Phil.)--Hong Kong University of Science and Technology, 2006. / Includes bibliographical references (leaves 81-83). Also available in electronic version.

Thesis (Ph.D.)--Hong Kong University of Science and Technology, 2007. / Includes bibliographical references. Also available in electronic version.

The aim of this thesis is to link perceived image quality to physical display parameters. This is done in the context of automotive displays. Specialities of automotive display applications like high ambient lighting conditions and the necessity to access information quickly are explained. A summary of readability models relevant to automotive applications is given and the difference between readability and perceived image quality is explained. The methodology chosen to investigate perceived image quality is the Image-Quality-Circle framework by Engeldrum (2000). Engeldrum states that observers form their image quality rating by weighting the visual attributes they perceive. Visual algorithms, which can be investigated via psychometric scaling, link visual attributes to the underlying physical image parameters which are typically measure by physical instruments. The visual attributes investigated in this thesis are perceived contrast, brightness, blackness and colourfulness. Perceived contrast, brightness and blackness are derived from display luminance via the DICOM just noticeable difference (JND) scale. Colourfulness is scaled based colour gamut in the CIE1931 chromaticity diagram. It was shown that image quality rating rises with growing perceived contrast; the limiting factors are glare and perceived blackness. In colourfulness scaling a linear relationship between colour gamut and colourfulness rating was demonstrated. Higher colourfulness can compensate lower brightness in perceived image quality.

No description available.

No description available.

After about five decades of extensive material research and device development, followed by massive investment in manufacturing technology, thin-film-transistor liquid-crystaldisplay (TFT-LCD) has finally become the dominant flat panel display technology. Nowadays, LCD performances, such as viewing angle, contrast ratio, and resolution, have reached acceptable levels. The remaining major technical challenges are response time, light efficiency, and sunlight readability. Fast response time is desired to reduce motion blur and to enable field sequential color displays using red (R), green (G), and blue (B) LEDs (light emitting diodes) without noticeable color breakup. Sequential RGB colors would eliminate the commonly used spatial color filters which in turn enhances light efficiency and resolution density by ~ 3X. In this dissertation, several new approaches for achieving fast-response LCDs are explored. From material viewpoint, the most straightforward approach for achieving fast response time is to employ a thin cell gap with high birefringence and low viscosity liquid crystal (LC). We investigated the thin cell approach theoretically and experimentally. Voltage shielding effect and anchoring energy effect of alignment layers are found to play important roles on operating voltage and response time. Simulations are carried out to understand the underlying physics and confirm the experimental results quantitatively. Another approach to realize fast response time is to explore novel device configuration. Here, we proposed a dual fringing-field switching (DFFS) mode in which small LC domains are iv formed following the distribution of fringing fields. Therefore, it exhibits submillisecond response time without using thin cell or overdrive/undershoot voltage method. The response time of the DFFS mode is ~20X faster than a conventional vertical aligned LCD. In addition, high optical efficiency is achieved from the complementary top and bottom active LC domains. Two transmissive and one transflective LCDs using DFFS mode are conceived and their electrooptical properties investigated. A shortcoming of DFFS LCDs is their fabrication complexity. To keep the advantages of this fast-response mode while avoiding the requirement of double-TFTs and pixel registration, we modified the device structure to transflective LCD which uses a single TFT in each pixel and vertical aligned positive dielectric anisotropy LC. Two types of electrodes are considered: fringing-field switching (FFS) and in-plane switching (IPS). Besides fast response time and high transmittance, such a transflective LCD shows good sunlight readability. As nematic LC is gradually approaching to its limit in term of response time, polymerstabilized blue phase (PSBP) LCD is emerging. It has potential to become next-generation display because of following revolutionary features: submillisecond response time, no need for alignment layer, good dark state and symmetric viewing angle, and cell gap insensitivity if IPS electrode is employed. In this dissertation, we studied the material-property correlation of Kerr effect-induced birefringence in nano-structured PSBP LC composites. Furthermore, a new device configuration of BP LCD with corrugated electrodes is proposed to solve two critical technical issues: high driving voltage and relatively low transmittance. The on-state voltage can be reduced from ~35 Vrms to ~10 Vrms which will enable TFT addressing, and the transmittance is improved from ~65% to ~85%. This new device configuration will accelerate the emergence of v PSBP LCD. Wide view is another important requirement for a high-end display. Several new LCD configurations with negative A-plate and biaxial plate as phase compensation films are proposed to achieve wide view and broadband operation. The underlying working principles are studied and detailed display performances are included in this dissertation.

No description available.

Display technology has developed significantly over the last decade, and it is becoming increasingly feasible to construct large, high-resolution displays. Prior work has shown a number of key performance advantages for these displays that can largely be attributed to the replacement of virtual navigation (e.g., panning and zooming) with physical navigation (e.g., moving, turning, glancing). This research moves beyond the question of performance or efficiency and examines ways in which the large, high-resolution display can support the cognitive demanding task of sensemaking. The core contribution of this work is to show that the physical properties of large, high- resolution displays create a fundamentally different environment from conventional displays, one that is inherently spatial, resulting in support for a greater range of embodied resources. To support this, we describe a series of studies that examined the process of sensemaking on one of these displays. These studies illustrate how the display becomes a cognitive partner of the the analyst, encouraging the use of the space for the externalization of the analyst's thought process or findings. We particularly highlight how the flexibility of the space sup- ports the use of incremental formalism, a process of gradually structuring information as understanding grows. Building on these observations, we have developed a new sensemaking environment called Analyst's Workspace (AW), which makes use of a large, high-resolution display as a core component of its design. The primary goal of AW is to provide an environment that unifies the activities of foraging and synthesis into a single investigative thread. AW addresses this goal through the use of an integrated spatial environment in which full text documents serve as primary sources of information, investigative tools for pursuing leads, and sensemaking artifacts that can be arranged in the space to encode information about relationships between events and entities. This work also provides a collection of design principles that fell out of the development of AW, and that we hope can guide future development of analytic tools on large, high-resolution displays. / Ph. D.

Geospatial intelligence analysts, epidemiologists, sociologists, and biologists are all faced with trying to understand massive datasets that require integrating spatial and multidimensional data. Information visualizations are often used to aid these scientists, but designing the visualizations is challenging. One aspect of the visualization design space is a choice of when to use a single complex integrated view and when to use multiple simple views. Because of the many tradeoffs involved with this decision it is not always clear which design to use. Additionally, as the cost of display technologies continues to decrease, large, high resolution displays are gradually becoming a more viable option for single users. These large displays offer new opportunities for scaling up visualization to very large datasets. Visualizations that are visually scalable are able to effectively display large datasets in terms of both graphical scalability (the number of pixels required) and perceptual scalability (the effectiveness of a visualization, measured in terms of user performance, as the amount of data being visualized is scaled-up). The purpose of this research was to compare information visualization designs for integrating spatial and multidimensional data in terms of their visual scalability for large, high resolution displays. Toward that goal a hierarchical design space was articulated and a series of user experiments were performed. A baseline was established by comparing user performance with opposing visualizations on a desktop monitor. Then, visualizations were compared as more information was added using the additional pixels available with a large, high resolution display. Results showed that integrated views were more visually scalable than multiple view visualizations. The visualizations tested were even scalable beyond the limits of visual acuity. User performance on certain tasks improved due to the additional information that was visualized even on a display with enough pixels to require physical navigation to visually distinguish all elements. The reasons for the benefits of integrated views on large, high resolution displays include a reduction in navigation due to spatial grouping and visual aggregation resulting in the emergence of patterns. These findings can help with the design of information visualizations for large, high resolution displays. / Ph. D.

Past research has found that infants between 1- and 4-months of age prefer extremely slow infant-directed (ID) speech to normal ID speech. Given that previous studies have paired the speech with a visual display, it was thought that perhaps the characteristics of the visual display affected infant preferences for rate of speaking. The present investigation was an attempt to explore this possibility. In Experiment 1, 3- to 4-month old infants were presented with ID-normal and ID-slow speech samples paired with displays of higher and lower complexity. Here, the results showed that infants preferred ID-slow to ID-normal speech regardless of display complexity. The purpose of Experiment 2 was to determine whether infant preference for speaking rate could be affected by structural differences in the visual displays. The same ID-normal and ID-slow speech samples used in Experiment 1 were paired with visual displays that either had components in a face-like or scrambled arrangement. The results of Experiment 2 showed that infants preferred the face-like display over the scrambled display regardless of what speech type was presented (i.e., there was not preference for ID-slow speech). The results of the study as a whole indicate that young infants prefer ID-slow to ID-normal speech, but that the presence of a face-like image may overshadow this preference. The study as a whole indicates that assertions about the power of ID-slow speech, as well as any auditory event in general, should not fail to address the potential influence of other sensory factors, particularly visual. / Ph. D.