Share Email Print

Proceedings Paper

Proof of principle for helmet-mounted display image quality tester
Author(s): Sheng-Jen Hsieh; Thomas H. Harding; Clarence E. Rash; Howard H. Beasley; John S. Martin
Format Member Price Non-Member Price
PDF $14.40 $18.00

Paper Abstract

Helmet mounted displays (HMDs) provide essential pilotage and fire control imagery information for pilots. To maintain system integrity and readiness, there is a need to develop an image quality-testing tool for HMDs. There is currently no such tool. A framework for development of an image quality tester for the Integrated Helmet and Display Sighting System (IHADSS) used in the U.S. Army's AH-64 was proposed in Hsieh et al. This paper presents the prototype development, summarizes the bench test findings using three IHADSS helmet display units (HDUs), and concludes with recommendations for future directions. The prototype consists of hardware (two cameras, sensors, image capture/data acquisition cards, battery pack, HDU holder, moveable rack and handle, and computer) and software algorithms for image capture and analysis. Two cameras with different apertures were mounted in parallel on a rack facing the HDU holder. A handle was designed to allow users to position the HDU in front of the two cameras. The HMD test pattern was then captured. Sensors are used to detect the position of the holder and whether the HDU was angled correctly in relation to the camera. Two sets of unified algorithms were designed to detect features presented by the two cameras. These features include focus, orientation, displacement, field-of-view, and number of gray-shades. Images of test pattern were captured and analyzed, and used to develop a specification for each inspection feature. Experiments were conducted to verify the robustness of the algorithms. A worst-case scenario for factors such as clock-wise and counterclockwise tilt, degree of focus, magnitude of brightness and contrast, and shifted images were set up and evaluated. Bench testing of three field-quality HDUs indicate that the image analysis algorithms are robust and able to detect the desired image features. Suggested future work includes development of a learning algorithm to automatically develop or revise feature specifications as the number of inspection samples increases.

Paper Details

Date Published: 8 September 2003
PDF: 12 pages
Proc. SPIE 5079, Helmet- and Head-Mounted Displays VIII: Technologies and Applications, (8 September 2003); doi: 10.1117/12.485548
Show Author Affiliations
Sheng-Jen Hsieh, Texas A&M Univ. (United States)
Thomas H. Harding, UES, Inc. (United States)
Clarence E. Rash, U.S. Army Aeromedical Research Lab. (United States)
Howard H. Beasley, UES, Inc. (United States)
John S. Martin, UES, Inc. (United States)

Published in SPIE Proceedings Vol. 5079:
Helmet- and Head-Mounted Displays VIII: Technologies and Applications
Clarence E. Rash; Colin E. Reese, Editor(s)

© SPIE. Terms of Use
Back to Top