Human Interface Technology Laboratory
University of Washington
The VRD functions by scanning light across the retina in a raster pattern. There is no screen within the display. It can be configured in various combinations of field of view and resolution. It has been configured to create images with details as small as 1 minute of arc, the limit of resolution of the eye. The VRD has RGB color and a broad brightness range for high contrast. Ultimately, the VRD's viewing characteristics should exceed those of any existing display. Furthermore, it can be configured as a see-through display, allowing superimposition of displayed images on those from the real world (augmented reality). The brightness of the VRD can be set so high that its images can be seen in regular daylight. The system uses laser light at extremely low power levels that are safe. The light intensity from the display is 1O4 below the maximal permissible exposure for laser light during normal operations in room light. Unlike any other scanner technology, the scanner system that the VRD is based on is extremely small and the scanner will be miniaturized to fit onto a light weight head mounted display.
The VRD display characteristics will allow it to be the basis of many unobtrusive display systems. For example, the brightness of the VRD in a augmented reality system will allow optics to be optimized for see through, rather than to see the displayed image. In an application such as interventional radiology, such a configuration would allow a physician to readily see the patient's body with images superimposed. This technology could be used assist in procedures such as needle biopsies. For such medical applications, high resolution, high contrast images with accurate color are needed. For applications such as virtual environments, the additional requirement of a field of view of at least 60 degrees while resolution is maintained is also necessary. A VRD based VR system could have a high resolution across a wide field of view, without being hampered by edges, distortions or other drawbacks of other display systems.
To improve the existing VRD technology, studies are needed on how the unique mode of illumination of the eye by the VRD affects its image properties. The research approach is based on an understanding of how the VRD functions compared to conventional displays. The mode of illumination of the retina by the VRD is quite different from that of a conventional, pixel based display. In an image that is made up of pixels, each pixel is turned on to a particular intensity and color. The whole area of the pixel is turned on simultaneously and its illumination persists 5 milliseconds or longer during the refresh cycle. In contrast, with the VRD a spot of light is swept across the retina. The illumination of a given retinal area is extremely brief (4O nanoseconds). The light used in the VRD is coherent and has a very narrow wavelength bandwidth. In the proposed experiments, as the image quality measures of resolution, contrast and flicker are measured, the sweep time, spot height and refresh rate will be varied. The vernier acuity and contrast sensitivity will be tested in users. Visual distortions will also be examined. At the earliest level of processing in the retina manipulations occur with the incoming that affect the perceived image. Resolution and contrast are enhanced. By manipulating the scanned beam characteristics so that beam optimally interacts with retinal processing, it should be possible to enhance the perceived brightness, contrast and resolution of an image from the VRD.
The VRD's mode of illumination affects the perception of color. Perceived brightness versus light power curves at wavelengths across the visible spectrum will be made to determine if the VRD saturates retinal photoreceptors. With different combinations of color sources and beam characteristics, effects on contrast, hue and saturation discrimination will be made. Particular combinations of colors may even make it possible to allow subjects with poor color perception to have enhanced use of a full color display.
[1] E.Viirre, R. Johnston, H. Pryor, and S. Nagata, "Laser safety analysis of a retinal scanning display system." Journal of Laser Applications, October, 1997
[2] M. Tidwell, A Virtual Retinal Display for Augmenting Ambient Visual Environments, Seattle WA.: University of Washington, 1995.
[3] L. E. Culham, F. W. Fitzke, G. T. Timberlake, and J. Marshall, "Use of scrolled text in a scanning laser ophthalmoscope to assess reading performance at different retinal locations," Ophthalmic Physiol Opt, vol. 12, pp. 281-6, 1992.
[4] G. Westheimer, "Spatial interaction in the human retina during scotopicvision," J Physiol Lond,, vol. 181, pp. 881-94, 1965.
[5] D. Y. Teller, C. Matter, W. D. Phillips, and K. Alexander, "Sensitization by annular surrounds: sensitization and masking," Vision Res, vol. 11, pp. 1445-58, 1971.
[6] S. Coren, "Retinal mechanisms in the perception of subjective contours:the contribution of lateral inhibition," Perception, vol. 20, pp. 181-91, 1991.
[7] V. C. Smith and J. Pokorny, "Chromatic-discrimination axes, CRT phosphor spectra, and individual variation in color vision," J Opt Soc Am A, vol. 12, pp. 27-35, 1995