A Virtual Retinal Display For Augmenting Ambient Visual Environments

by Michael Tidwell

[Previous Chapter][Table of Contents][Next Chapter]


Chapter 4: Survey of Helmet Mounted Displays

4.1 The Head Mounted Display and Virtual Images

A general method of classifying HMD's is by whether they are occluded or see-through displays. The occluded (or inclusive) display is one where only the image produced by the display is visible to the viewer. The see-through (or augmented vision) display is one where the viewer sees both the image produced by the display and the ambient scene.

Another method of classifying HMD's is by how many images are presented and to which eyes. The three classifications are monocular, biocular, and binocular. A monocular display presents one image to one eye. A biocular display presents one image to both eyes (i.e. both eyes see the same image). Finally, a binocular display presents different images to each eye. It is only with a binocular display that true stereoscopic images can be presented.

Head mounted displays, in general, project virtual images into the eye. Specifically, a see-through (or augmented vision) helmet mounted display presents a virtual image which is perceived as being in the same place as an ambient scene. Viewing a real image requires an image plane or viewing screen and defeats the purpose of a see-through display. The fundamental difference between a real and a virtual image is that a real image can be viewed at an accessible plane in space (with a screen of some sort) and a virtual image cannot. Figure 4.1 shows an example of real image formation by a lens and Figure 4.2 shows as example of virtual image formation by a mirror. Figure 4.3 illustrates a typical approach for forming a virtual image of a screen in a head mounted display.

Figure 4.1: Formation of a real image from a real object by a lens of focal length f.

Figure 4.2: Formation of a virtual image from a real object by a mirror.

Figure 4.3: A general approach for imaging CRT's in a see-through HMD.

4.2 Image Sources for Virtual Image Displays

Current helmet mounted displays typically use either a liquid crystal matrix array or cathode ray tube to generate the image. A transmission mode liquid crystal display is an electrically addressed two dimensional matrix of holding cells filled with a liquid crystal material. When the cell is switched either on (high voltage) or off (low voltage), depending on the type of liquid crystal cell, the transmittance of light through the cell changes. Video switching of the individual cells creates each pixel of a video display. Liquid crystal cells can be covered by a mask which transmits light of a narrow frequency band. The combination of three cells (red, green, and blue) constitutes a color pixel. The cells of a liquid crystal matrix are addressable and capable of producing different colors of light and are suitable devices through which a video image can be produced. Most liquid crystal displays (LCD's) for helmet mounted display applications are on the order of 0.7 [in.] diagonally with 340 by 230 elements. LCD's have also been developed one inch by one inch in size with up to 640 by 480 elements in the array. Therefore, to create a large field of view image desired in helmet mounted displays, the LCD screen must be significantly magnified or placed close to the eye. Large angular magnification causes the angular extent on the retina of each liquid crystal cell to increase. Even though a large angular magnification makes the picture appear larger, there is a significant reduction in the perceived resolution or detail in the display for large angular magnifications.

An alternate technology for display generation is the cathode ray tube (CRT). The cathode ray tube operates by modulating, accelerating, and deflecting an electron beam through a vacuum onto a phosphor face plate. The electron beam strikes cells of phosphorescent materials on a screen causing them to glow. Different phosphors glow at different wavelengths to give a color display. Deflection electronics control the beam location in synchronization with a video signal to create a video image. Typical CRT's used in helmet mounted displays are about one inch in diameter with up to 1000 by 1000 elements. It is, however, difficult to create a full color CRT of this size. Again, to create a large field of view image, the screen must be magnified and/or placed close to the eye.

Some head mounted displays consist of CRT's imaged into a fiber optic bundle which carries the signal to the eye. The fiber optic ray bundle approach offers no improvement in image quality but can allow for better ergonomics on the head. The collection of fibers acts as a secondary screen which must then be imaged into the eye. For a large field of view display, the fiber bundle will have to be magnified or placed close to the eye.

4.3 Matrix of Currently Available Head Mounted Displays.

The following helmet mounted display systems are commercially available or under development [49,22].

Table IV.1. Current head mounted displays.

Company/

Model

Display

Technology

Field of

View [deg.]

(horiz. x vert.)

Overlap Pixel Size

[arc min.]

Cost

[U.S. dollar]

Virtual I/OCRT's w/

Color

Rendering

40 30
Not

Available
3.75
40, 000
Astounding

Technologies /2001

Video Visor

Active matrix

LCD's

30 22.5
100 [%]
4.21
795
CAE-Electronics

/Fiber Optic HMD

CRT's projector

through fiber optics

72.5 55

(bckgnd.)

24 x 18

(inset)
25 [deg.]
6.0 (bckgnd.)
By Quote Only
CAE-Link

/Clear Vue

Mono-chrome CRT's w/ LCD filters
60 x 35
20 [deg.]
2.4
65,000
FORTE Technolo-gies /VFX 1Active matrix

LCD's

46.4 35.2
Adjustable
4.43
<1, 000
Johns Hopkins Low Vision Enhancement System 0.6 inch b&w CRT's
50 37.5
100 [%]
4.68
5,200
Kaiser Electro-Optics /SIM EYE 40 Mono or field sequential

CRT's

60 40
40 [deg.]
2.81
145,000
Kaiser Electro-Optics /500 pv VIM Active matrix

LCD's

40 30
100 [%]
3.38
2,975
Kaiser Electro-Optics /1000 pv VIM Active matrix

LCD's

100 30
36 [deg.]
8.05
9,975
General Reality /Cyber EyeActive matrix color LCD's
22.4 16.8
100 [%]
3.2
2,49

stereo

1,995 mono
Liquid Image /MRG 2Active matrix

color

LCD

84 65
Single

screen
7.0
3,495
Liquid Image /MRG 2Active matrix color LCD
60 46
Single screen
7.52
2,199
n-Vision /Datavisor 9cCRT's and

LCD shutter

devices

50 37
0 to 100 [%]
1.98
40,000 -

50,000
Polhemus Laboratories /Looking Glass CRT's or LCD's through fiber optics
35 - 60 [H]

35 - 50 [V]
20 - 80 [%]
1.5 to 3.0
55,000
RPI /HMSI Micro Model 900LCD
65 40
100 [%]
5.0
3,260
RPI /HMSI Multi User Model 950LCD
100 60
100 [%]
4.0
7,500
RPI /HMSI OEM 1200LCD
80 50
100 [%]
3.0
300

(OEM only)
VictorMaxx Technologies /Cyber Maxx Active

matrix LCD's

62 54
100 [%]
7.75
699
Virtual Reality /Hi-Resolution Mono-chrome Personal Immersive Display 131 CRT's
40 - 60 [H]

30 [v]
0 to 20 [deg.]
1.875 to 2.82
47,000
Virtual Reality /Personal Immersive Display P1 LCD's
62 30
31 [deg.]
7.76
5,600
U.S. Air Force Human Resources Lab LCD and CRT versions

through fiber optics

160 80

(backgnd.)

inset N/A
N/A
1.5 (inset)
Experi-mental
Virtual Research /EYEGEN 3Mono-chrome CRT's w/

color wheels

36 24.3
Custom w/

55, 70, 85, 100 [%] options
6.5
7,900
Virtual Research /VR 4 Active matrix LCD's
48 36
100 [%]
8.1
7,900
Virtuality Entertainment VISETTE 2 LCD
60 46.87
N/A
4.76
6, 000
Vista Controls /See-Through Armor Active matrix LCD
40 30
100 [%]
3.75
25,000

4.4 Disadvantages of Current Technologies

Some obvious disadvantages to current technology are evident from the preceding table. Cost is a clear disadvantage in some displays as miniature cathode ray tubes can alone cost thousands of dollars. In general, liquid crystal display based systems are cheaper. LCD screens suffer from low light output, however, and are usually not suitable for see-through displays because see-through displays inherently require higher intensity image sources. Weight and size are of great concern in most head mounted display applications. CRT and LCD screen based technologies are usually bulky, heavy, and produce substantial torque on the head-spine system. In some CRT and LCD displays, the screens themselves are bulky and sometimes heavy (e.g. approximately 1-2 pounds for some CRT's). In most CRT and LCD HMD's, the relay and collimation optics are bulky and heavy, especially in the case of the CRT. If a screen is made small for lighter weight, the screen either must be placed closer to the eye or magnified more to achieve a large field of view. A closer relation to the eye degrades eye relief, the space between the eye and the final element of the system) and greater magnification degrades resolution. The inherent trade-off between resolution, eye relief, screen size, and weight in screen based HMD's is a severely limiting aspect of their nature.


Human Interface Technology Laboratory