Academic journal article Human Factors

Image Quality of Two-Primary Color Active-Matrix Liquid-Crystal Displays

Academic journal article Human Factors

Image Quality of Two-Primary Color Active-Matrix Liquid-Crystal Displays

Article excerpt


Most helmet-mounted displays (HMDs) designed for use in military aircraft use a miniature monochrome CRT as the image source. Pilots would like a color image source because color can enhance perceptual segregation of symbols, reinforce shape coding (e.g., friendly vs. hostile vs. unknown), and help segregate symbols from forward-looking infrared (FLIR) images (Post, 1992). However, pilots are reluctant to sacrifice luminance or resolution to obtain color.

Recently the subtractive-color active-matrix liquid-crystal display (AMLCD) emerged as a near-term technology that can provide color without these sacrifices and can meet the other demands of daytime airborne HMDs (Post et al., 1994). For this reason, subtractive-color AML-CDs have been adopted as HMD image sources in the U.S. Army's Advanced Visionics System program (Kanahele & Buckanin, 1996) and the U.S. Air Force's Helmet-Mounted Sight Plus (HMS+) program. Readers desiring a detailed analysis and comparison of subtractive versus alternative color display technologies for HMDs may consult Post et al. (1994), which also explains why subtractive-color AMLCDs are not well suited for direct-view applications.

HMS+ is intended mainly to support weapon aiming and the display of conventional head-up display symbology, though display of FLIR images with overlapping symbology is also planned. Hence only a few colors for simple segregation and coding purposes are thought to be needed. For this reason, a two-primary, red-plus-green display (which can produce all hues between red and green - e.g., orange and yellow - as well as red and green) has been specified for HMS+ rather than a full-color display. For a subtractive-color AMLCD, the transmittance of a two-primary display is at least two to four times that of an otherwise equivalent full-color display, and the manufacturing cost is 39% lower (Franklin & Reinhart, 1997). Thus the use of two-primary color for HMS+ will increase luminous efficacy and reduce cost significantly.

The main reason subtractive-color AMLCDs are more attractive than conventional color displays for HMDs has to do with resolution. On one hand, a conventional color display produces its color gamut by presenting red, green, and blue subpixels that are integrated spatially at the eye to produce what appear to be full-color pixels. Thus the pixels are larger than the subpixels that constitute them. A subtractive-color AMLCD, on the other hand, consists of a stack of spatially registered monochrome AMLCDs, each of which controls the red, green, or blue portion of the light traversing the stack. In this design the subpixels are layered physically, so each pixel is the same size as the subpixels constituting it [ILLUSTRATION FOR FIGURE 1 OMITTED]. Thus a subtractive-color AMLCD uses the same color-generating principle as a color photograph.

For a given subpixel size and display area, a subtractive-color AMLCD has many more pixels and, hence, higher resolution than does an integrative-color display. In principle, an integratire display's subpixel density can be increased to match a subtractive display's resolution, but this requires the use of smaller subpixels, which, in the case of an AMLCD, implies a lower transmittance. Therefore, it has been assumed that subtractive-color AMLCDs provide better image quality than integrative-color AMLCDs, even if resolution is held constant. However, this assumed superiority has never been demonstrated empirically.

A disadvantage of subtractive color is the fact that light for each pixel must pass through an aperture at each layer in the stack and therefore diffracts at each layer. Thus light that is meant to exit a given pixel in the final, front layer tends to spread into adjacent pixels as it traverses the stack, which reduces contrast and produces color shifts. Diffraction can be modeled numerically as a function of pixel size and shape, but, again, its impact on subjective image quality has not been assessed. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.