H. Lee Task

Night vision goggle visual acuity assessment: results of an interagency test

There are several parameters that are used to characterize the quality of a night vision goggle (NVG) such as resolution, gain, field-of-view, visual acuity, etc. One of the primary parameters is visual acuity or resolution of the NVG. These two terms are often used interchangeably primarily because of the measurement methods employed. The objectives of this paper are to present: (1) an argument as to why NVG visual acuity and resolution should be considered as distinctly different parameters, (2) descriptions of different methods of measuring visual acuity and resolution, and (3) the results of a blind test by several agencies to measure the resolution of the same two NVGs (four oculars).

Theoretical and applied aspects of night vision goggle resolution and visual acuity assessment

The image quality of night vision goggles is often expressed in terms of visual acuity, resolution or modulation transfer function. The primary reason for providing a measure of image quality is the underlying assumption that the image quality metric correlates with the level of visual performance that one could expect when using the device, for example, target detection or target recognition performance. This paper provides a theoretical analysis of the relationships between these three image quality metrics: visual acuity, resolution and modulation transfer function. Results from laboratory and field studies were used to relate these metrics to visual performance. These results can also be applied to non-image intensifier based imaging systems such as a helmet-mounted display coupled to an imaging sensor.

The impact of target luminance and radiance on night vision device visual performance testing

Visual performance through night-vision devices (NVDs) is a function of many parameters such as target contrast, objective and eyepiece lens focus, signal/noise of the image intensifier tube, quality of the image intensifier, night-vision goggle (NVG) gain, and NVG output luminance to the eye. The NVG output luminance depends on the NVG sensitive radiance emitted (or reflected) from the visual acuity target (usually a vision testing chart). The primary topic of this paper is the standardization (or lack thereof) of the radiance levels used for NVG visual acuity testing. The visual acuity chart light level might be determined in either photometric (luminance) units or radiometric (radiance) units. The light levels are often described as “starlight,” “quarter moon,” or “optimum” light levels and may not actually provide any quantitative photometric or radiometric information. While these terms may be useful to pilots and the users of night-vision devices, they are inadequate for accurate visual performance testing. This is because there is no widely accepted agreement in the night vision community as to the radiance or luminance level of the target that corresponds to the various named light levels. This paper examines the range of values for “starlight,” “quarter moon,” and “optimum” light commonly used by the night vision community and referenced in the literature. The impact on performance testing of variations in target luminance/radiance levels is also examined. Arguments for standardizing on NVG-weighted radiometric units for testing night-vision devices instead of photometric units are presented. In addition, the differences between theoretical weighted radiance and actual weighted radiance are also discussed.

Effect on vision of light scatter from HMD visors and aircraft windscreens

The amount of scattered light, or haze, typically increases as transparent materials age, wear, become dirty, or become scratched from cleaning. Light scattered from scratched aircraft transparencies, such as windscreen, head-up-display combiners, and helmet visors, can potentially reduce pilot visual performance and reduce target detection range. Presented in this paper are the results of an investigation of light scattered from transparencies exhibiting different levels of wear and surface damage. Two methods of measuring scattered light are compared. Visual performance under conditions of white light scatter relevant to the use of helmet-mounted displays in the cockpit is also examined.

Effects of aircraft windscreen on helmet-mounted display/tracker aiming accuracy

Modern fighter aircraft windscreens are typically made of curved, transparent plastic for improved aero-dynamics and bird-strike protection. Since they are curved these transparencies often refract light in such a way that a pilot looking through the transparency will see a target in a location other than where it really is. This effect has been known for many years and methods to correct the aircraft head-up display (HUD) for these angular deviations have been developed and employed. The same problem will occur for helmet-mounted displays (HMDs) used for target acquisition only worse due to the fact the pilot can look through any part of the transparency instead of being constrained to just the forward section as in the case of the HUD. To determine the potential impact of these windscreen refraction errors two F-15 windscreens were measured; one acrylic and one multilayer acrylic and polycarbonate laminate. The average aiming error measured for the acrylic was 3.6 milliradians with a maximum error of 9.0 milliradians. The laminated windscreen was slightly worse at 4.1 milliradians average error and 10.5 milliradians maximum. These aiming errors were greatly reduced by employing correction algorithms which could be applied to the aiming information on the HMD. Subtleties of coordinate systems and roll correction are also addressed.

Development of a dichoptic foveal/peripheral head-mounted display with partial binocular overlap

Previous foveal/peripheral display systems have typically combined the foveal and peripheral views optically, in a single eye, in order to provide simultaneously both high resolution and wide field of view from a limited number of pixels. While quite effective, this approach can lead to cumbersome optical designs that are not well suited to head-mounted displays. A simpler approach may be possible in the form of a dichoptic vision system, wherein each eye receives a different field of view (FOV) of the same scene, at different resolutions. One eye would be presented with highresolution narrow-FOV foveal imagery, while the other would receive a much wider peripheral FOV. Binocular overlap in the central region would provide some degree of stereoscopic depth perception. It remains to be determined, however, if such a system would be acceptable to users, or if binocular rivalry or other adverse side-effects would degrade visual task performance compared to conventional head-mounted binocular displays. In this paper, we describe a preliminary dichoptic foveal/peripheral vision system and suggest methods by which its usability and performance can be assessed. This effort was funded by the U.S. Air Force Research Laboratory Human Performance Wing under SBIR Topic AF093-018.

Measurement of visual performance through scattering visors and aerospace transparencies

Light scattered from helmet visors and aerospace transparencies is known to reduce visual performance. One popular measurement technique, maintained by the American Society for Testing and Materials, is ASTM D 1003. It is a standard procedure used to measure haze inherent in transparent materials, which is defined as the percent of the total transmitted light that is scattered. However, research has shown that visual acuity measured through several different types of helmet visors does not correlate well with visor haze. This is most likely due to the fact that the amount of light scattered from a transparent material depends heavily on the light illuminating the transparency and on the viewing geometry, behavior that ASTM D 1003 does not characterized. Scattered light causes transparent parts to appear luminescent and imparts a veiling luminance when superimposed over a target, reducing target contrast and inducing a visual performance loss. This paper describes an experiment in which threshold target background luminance, the luminance at which a target was barely visible, was measured for a number of observers viewing a Landolt C target through several levels of veiling luminance. Threshold luminance was examined for predictable behavior with respect to veiling luminance.

Measurement of military helmet- and head-mounted display (HMD) visor optical properties

This paper examines the light transmission, reflection, and scattering characteristics of military helmet visors used for see-through helmet-mounted displays (HMDs). HMDs used for the within-visual-range counter-air mission normally use the inner surface of the helmet visor to reflect the HMD image to the pilots eye. This approach is popular because it minimizes any optical structures that interfere with the pilots vision, while also maximizing see-through to the ambient scene. In most cases, a reflective coating, which increases the cost of the helmet visor significantly, must be applied in the inner surfaces in order to achieve enough contrast between the HMD image and the external light passing through the visor. Recently, with the development of high luminance miniature cathode-ray-tubes, it has been possible to eliminate the reflective coatings on neutral density helmet visors having a see-through range of 13 - 35%. This paper examines the light management properties of both types of visors. The paper stresses measurement techniques that produce repeatable results and what these results might imply about visual performance under operational lighting conditions.

A comparison of sensor resolution assessment by human vision versus custom software for Landolt C and triangle resolution targets

This paper is the fifth in a series exploring the possibility of using a synthetic observer to assess the resolution of both real and synthetic (fused) sensors. The previous paper introduced an Automatic Triangle Orientation Detection Algorithm (ATODA) that was capable of recognizing the orientation of an equilateral triangle used as a resolution target, which complemented the Automatic Landolt C Orientation Recognition (ALCOR) software developed earlier. Three different spectral band sensors (infrared, near infrared and visible) were used to collect images that included both resolution targets and militarily relevant targets at multiple distances. The resolution targets were evaluated using the two software algorithms described above. For the current study, subjects viewed the same set of images previously used in order to obtain human-based assessments of the resolutions of these three sensors for comparison with the automated approaches. In addition, the same set of images contained hand-held target objects so that human performance in recognizing the targets could be compared to both the automated and human-based assessment of resolution for each sensor.

Object recognition methodology for the assessment of multi-spectral fusion algorithms: phase 1

In this effort we acquired and registered a multi-spectral dynamic image test set with the intent of using the imagery to assess the operational effectiveness of static and dynamic image fusion techniques for a range of relevant military tasks. This paper describes the image acquisition methodology, the planned human visual performance task approach, the lessons learned during image acquisition and the plans for a future, improved image set, resolution assessment methodology and human visual performance task.