U.S. flag An official website of the United States government

U.S. Food and Drug Administration

Contrast Measurement Method for Optical See-Through Augmented Reality Head-Mounted Displays

Catalog of Regulatory Science Tools to Help Assess New Medical Devices 

 

This regulatory science tool (RST) describes a method to characterize digital image quality on optical see-through augmented reality head-mounted displays (AR HMDs) in ambient lighting conditions. The method can be implemented to measure AR display performance, such as Michelson contrast and contrast uniformity.

 

Technical Description

The input to this RST is a set of checkerboard patterns that should be rendered on an evaluated AR HMD in different ambient light conditions. This RST specifies the test pattern, experimental setup, and image acquisition and processing procedures for display contrast measurements. Technical description of the tool is provided below:

  1. Test pattern: As shown in Figure 1(a), checkerboard pattern with a combination of display gray levels that span the display dynamic range should be prepared prior to the test. See detail described in a peer-reviewed publication [1] and Appendix A1. The test pattern shall be rendered on the headset and centered with respect to the optical axis of the HMD.
    To assist with this, the FDA has released an RST “Toolkit for Evaluation of Head Mounted Display Image Quality” based on the WebXR platform [2]. The WebXR toolkit can be used to prepare and render the checkerboard patterns on an HMD with a compatible web browser.
  2. Experimental setup: As shown in Figure 1(b), this method uses a luminance-calibrated light measuring device (LMD), such as an imaging photometer, that provides a wide-view spatial luminance measurement (see Appendix A2 for technical requirements on the LMD [3, 4] ).
    To test the image quality of AR HMD with optical see-through, a uniform ambient light field, such as using an integrating sphere as the light source, can be overlaid on the digital test pattern [1].
  3. Image acquisition: 2D luminance profiles of the test patterns should be acquired using the LMD. The measurement shall be repeated using different luminance of the ambient light (e.g., three luminance levels from dark to a flat-field luminance corresponding to the intended use of the device). Detailed step-by-step implementation of this tool for image acquisition is described in Appendix A3.
  4. Analysis: Appendix A4 provides detailed instructions on image processing procedures to measure the AR HMD luminance and contrast characteristics. The output of this tool includes AR contrast for various display gray level combinations and contrast uniformity maps [1].
Figure 1 ChuminZhao Contrast Method
Figure 1. (a) Illustration of checkerboard pattern for contrast measurements between gray levels {IL, IH} at multiple ROIs centered at {x, y}. (b) Experimental setup for AR display characterization [1]. 

Intended Purpose

This tool provides a method to measure digital contrast in optical see-through AR head-mounted displays. The test method can be implemented by both medical device developers and testing labs for bench testing of the digital image quality for optical see-through augmented reality HMDs.

Testing

The method has been tested and validated on two AR HMDs – the Microsoft HoloLens 2 and Moverio BT-300 [1]. The tests involve repeating the measurement using different gray levels and ambient conditions. The results demonstrate that the method is reproducible on these two HMDs with different hardware, software, and rendering techniques [1].

Limitations

The contrast measurement takes a wide-view image that emulates a pupil rotation scheme using a static light measuring device (LMD), which is corresponding to the pupil rotation geometry as described in IEC 63145-20-10 [3]. The wide-view setup takes a single measurement to compute the Michelson contrast across the evaluated field of view but may introduce additional aberration at the periphery away from the optical axis of the LMD. Alternatively, the contrast measurement method can also be implemented in an eye rotation setup by repeating the contrast measurement at each target (gaze) location in accordance with the eye rotation geometry described in IEC 63145-20-10 [3].
This RST is applied to augmented reality (AR) devices with optical see-through content. The method can also be potentially implemented to evaluated video see-through headsets. Further evaluation of the contrast measurement method for video see-through or other types of AR devices is needed in future work.

Supporting Documentation

The method with detailed instructions and validation is described in the following publication:

  1. Zhao, Chumin, Ryan Beams, Matthew Johnson, and Aldo Badano. "18‐2: Assessment of Image Quality in Augmented Reality Displays Using a Computational Model of Target Detectability." In SID Symposium Digest of Technical Papers, vol. 53, no. 1, pp. 194-197. 2022. https://doi.org/10.1002/sdtp.15451

  2. FDA Regulatory Science Tool: “Toolkit for Evaluation of Head Mounted Display Image Quality.”

    • The WebXR toolkit is provided on GitHub

    Reference Standard Documents

  3. IEC 63145-20-10:2019 Eyewear display - Part 20-10: Fundamental measurement methods - Optical properties.
  4. https://www.sid.org/Standards/ICDM#8271500-icdm-info.
  5. IEC 63145-20-20:2019 Eyewear display - Part 20-20: Fundamental measurement methods - Image quality.

Contact

Tool Reference

  • In addition to citing relevant publications please reference the use of this tool using RST24MX04.01

For more information