Open Access
Add to BibSonomy
Abstract
We present an experimental method that utilizes a conic probe attached to an integrating sphere for characterizing the small-spot luminance of virtual reality (VR) head-mounted displays (HMDs). This allows for the measurement of relative luminance in virtual scenes and the assessment of VR displays for visualizing grayscale medical images. We characterized the relative luminance of the VIVE Pro and Oculus Rift by displaying test patterns with a circular dark spot of varying diameter in the bright field and compared the probe performance with commercially available systems. Compared to a conventional flat-panel display, the results show a significant level of veiling glare in the HMDs using both a customized conic probe with improved optical performance and an advanced imaging photometer. We also found no significant change in the relative luminance response with the probe’s working distance for the HMDs due to their unique optical architecture for near-eye viewing conditions. Our results highlight the issue of high veiling glare in HMDs and the importance of small-spot luminance measurements to adequately evaluate the technical performance of HMDs for emerging medical applications.
© 2023 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Virtual reality (VR) head-mounted displays (HMDs) are increasingly used for various medical applications, including therapy [1,2], rehabilitation [3], surgery planning [4], and diagnostics [5,6]. Medical VR headsets also have the potential to provide immediate access to diagnostic imaging results that do not rely on diagnostic and review workstations [7] in emergency use scenarios, such as the COVID-19 pandemic [8]. A high volume of imaging exams with the surge of patients in a pandemic may strain image interpretation workflow in limited workstations [9]. HMDs might provide an appropriate visualization solution to display diagnostic images from patients’ records, including chest x-rays and CT datasets [10,11]. Accurate luminance in virtual scenes is a key consideration for these emerging applications, especially for solutions that require visualization and interpretation of grayscale medical images, where luminance may convey diagnostic information. Recent studies have begun to evaluate luminance nonuniformity and discomfort luminance levels in various types of HMDs [12,13].
Since luminance characteristics of a display affect image quality, medical display performance with respect to luminance has been extensively studied in traditional displays, such as flat-panel liquid crystal display (LCD) monitors that offer high dynamic range and fine spatial resolution [14–17]. High dynamic range VR displays are also currently being actively developed and implemented in advanced HMDs [18–20]. An increase in the luminance level of dark regions in the presence of bright scenes in the field of view is known to degrade image quality by reducing perceived contrast [21]. This image degradation is associated with veiling glare due to the diffuse spread of unwanted light in the image plane. For a detection task, the contrast reduction due to veiling glare is known to hinder the detectability of low-contrast targets that are clinically relevant [22–24]. In chest x-ray images for COVID-19, this would mean affecting the detectability of lung opacities associated with pneumonia. Therefore, the optical properties of medical-grade displays are designed to minimize contamination of display luminance and maintain image contrast in the low luminance range [25,26]. However, the performance standards and compliance levels of commercial headsets for grayscale display have not been established, suggesting the need to focus on characterizing HMDs’ luminance using suitable methods for near-eye displays [27–29]. The measurement requirements for HMDs are different from those for flat-panel displays because HMDs create virtual images using optics for near-eye viewing conditions designed for small entrance pupils [27]. Additional optics in HMDs can cause increased light scattering and glare compared to traditional flat-panel displays. By comparing their performance, we can better understand veiling glare issues in HMDs and assess their potential as medical display devices.
In this study, we report an experimental method that utilizes a custom conic probe attached to an integrating sphere for characterizing the small-spot luminance of HMDs. The conic probe is designed to minimize signal contamination from bright regions outside the measurement spot and accurately measure relative luminance in virtual scenes with a wide range of luminance values. The relative luminance is characterized in two HMDs at three working distances by displaying bright circular test patterns with a central dark spot of varying diameter, where the relative luminance is the ratio of the average luminance values of the dark spot to the white circle.
2. Methods
2.1 Custom probe design
To measure luminance from small spots with minimal contamination from surrounding bright regions, we designed a conic probe (black plastic) with three collimating apertures around 4 mm in diameter, mimicking the typical size of the human entrance pupil [30,31].
A similar collimated design was implemented before using multiple baffles to measure veiling glare in a traditional display with a probe entrance aperture of 9 mm [24]. For light entering at oblique angles, the conic geometry of the probe with baffles was reported to minimize unwanted scattered light reaching the detector or the display surface in the measurement spot [24]. In our design, one end of the conic probe was externally threaded (SM05) to attach to the entrance port of an integrating sphere (IS200-4, Thorlabs, USA) via a ring-actuated iris diaphragm (SM05D5D, Thorlabs, USA). Since the aperture of the iris needed to be manually adjusted and locked in, the variability in the reported iris diameter was $\pm$0.6 mm. The schematic of the conic probe attached to an integrating sphere is shown in Fig. 1. The probe entrance aperture was 4.3 mm and the angular FOV was around 3$^{\circ }$. We used an integrating sphere with top and back ports (12.5 mm diameter) for successively capturing the FOV image and measuring the relative luminance without changing the measurement setup. There were no baffles inside the integrating sphere. A collimated response of the probe is illustrated by the rays of light passing through the conic probe in Fig. 1. When the light reaches the integrating sphere at the back port, it undergoes multiple diffuse reflections to get evenly distributed inside the sphere if the back port is closed. The principal rays reaching the back port define the FOV of the probe.
Fig. 1. Schematic of the custom conic probe with three collimating apertures attached to an integrating sphere. Collimated rays of light (dark red line) from a measurement spot pass through the conic probe and reach the back port of an integrating sphere. Light can then go through multiple diffuse reflections (red arrow) to get uniformly distributed inside the integrating sphere.
2.2 Experimental setup
Figure 2 shows the experimental setup using the custom probe for characterizing the luminance of VR HMDs. The conic probe was aligned along the optical axis of the lens at its center for the close contact measurements. The distance from the conic probe tip to the HMD lens was 0.5 mm. A monochrome camera (FLIR Blackfly, BFS-U3-122S6M-C, Teledyne FLIR, USA) was used for the alignment process and to capture the FOV images from the back port of the integrating sphere.
Fig. 2. Experimental setup using the custom probe for characterizing HMDs luminance, where the conic probe is pointing to the center of the VR lens. The monochrome camera is used to capture the FOV image, the silicon photodiode with photopic filter and research radiometer to measure luminance values, and the translation stage to change the working distance of the probe.
Luminance was measured either from the top or back port using a silicon photodiode sensor (33 mm$^2$ active area) with a photopic filter (SED033/Y, International Light Technologies, USA) connected to a research radiometer (ILT5000, International Light Technologies, USA). For the top port measurement, we first captured the FOV images using a monochrome camera then closed the backport for luminance measurement. For the back port luminance measurement in the FOV of the probe, we aligned the probe tip to the lens center, captured the FOV images, and removed the camera to attach the detector in the backport without moving the setup. One-axis translation stage was used to change the working distance of the probe from 0.5 mm to 20 mm. VIVE Pro (HTC Corporation, Taiwan) and Oculus Rift (Meta, USA) were two HMDs evaluated in this study using our custom probe.
The results were compared with three commercially available luminance measurement systems: a luminance detector (SED033/Y/R, International Light Technologies, USA), an imaging photometer (LMK6-12, TechnoTeam Vision, Germany), and a luminance pen probe (SPD025Y, International Light Technologies, USA). The SED033/Y/R luminance detector has a 1.5$^{\circ }$ FOV and it covers 25.4 mm spot size on contact to measure average luminance. The LMK6-12 imaging photometer is capable of mapping a small spot luminance (about 1 mm$^2$ circular field) across a large FOV of $\geq$ 60$^{\circ }$ and has a camera system with low internal glare. The SPD025Y luminance pen probe offers to measure average luminance from 5 mm spot size on contact with a 14$^{\circ }$ FOV. The luminance pen probe was used for reference measurements on a laptop display (13.3-inch IPS LCD with 1920 x 1080 screen resolution, Hewlett-Packard, USA) that showed typical small spot luminance characteristics of a conventional flat panel display. It should be noted that since the custom probe also exhibited similar behavior for reference measurement, we only used one set of data for the laptop display. In addition, the luminance pen probe data for the HMDs were not included because they showed poor small spot measurement performance than the custom probe.
2.3 Relative luminance measurement
To characterize small spot luminance measurement performance of HMDs, we measured TG18-like test patterns [26] generated in MATLAB (Mathworks, USA). As shown in Fig. 3, the test patterns consist of a central dark spot of varying diameters in a large bright circular field (white circle) of 50 mm fixed diameter. The diameter of the dark spot was varied up to the size of the white circle. These test patterns were displayed in the VIVE Pro and Oculus Rift using SteamVR (Valve Corporation, USA). We obtained a bright-field luminance of around 133 and 80 cd/m$^2$ for the VIVE Pro and Oculus Rift, respectively. The relative luminance was calculated as a ratio of the average luminance values of the dark spot to the white circle. The white circle measurement was performed by displaying a test pattern with a dark spot diameter of 0, making the relative luminance for a bright region equal to 1. All measurements were performed in the darkroom at three working distances of 0.5, 10, and 20 mm. We performed a total of four measurements per HMD with two repeated measurements in the right and left lens for each test pattern.
Fig. 3. Test patterns for small spot luminance characterization showing bright circular test patterns with a central dark spot of varying diameter from 0.5 to 50 mm. The bright circular field diameter is fixed at 50 mm.
3. Results and discussions
Figure 4 shows the representative FOV images of the displayed test patterns in the two HMDs. The FOV image was captured from the back port of the integrating sphere in our experimental setup for close contact measurements (Fig. 2). The distance between the conic probe tip and the HMD lens defining the working distance was 0.5 mm. In the captured images (Fig. 4), the dark spots are approximately in the center of the FOV, and their size increases accordingly to cover the FOV of the probe when we display the test patterns with the dark spot diameter varying from 0 to 2.5 mm. The result shows that the test patterns are appropriately displayed in the HMDs, and the collimated probe is aligned to point to a dark spot. The 2.5 mm dark spots cover most of the FOV image without a significant presence of the surrounding white region. Beyond 2.5 mm, we observed that the dark spots $\geq$ 3.3 mm completely covered the FOV image, yielding a dark image for both the VIVE Pro and Oculus Rift. This observation is consistent with the small FOV of the probe for image capture determined by the size of the iris around 3.8 mm $\pm$ 0.6 mm in the entrance port of the integrating sphere (Fig. 1).
Fig. 4. Representative FOV images captured using our experimental setup after the alignment process and displaying the test patterns in the VR headsets.
After capturing the FOV images, we characterized the relative luminance of the VIVE Pro and Oculus Rift with test patterns and compared the probe performance with the commercially available systems for close contact measurements at a 0.5 mm working distance. The relative luminance of the HMDs from the custom probe was measured from the top port of the integrating sphere (Fig. 2). Figure 5 shows significant signal contamination from the bright regions of the displayed test patterns in the HMDs for all the measured dark spots, irrespective of the dark spot size and the luminance measurement system used. In general for the HMDs, the relative luminance as a function of the dark spot diameter does not decrease as quickly as the reference measurement from a laptop display that reaches a background level of 1.3 x 10$^{-3}$. The measured relative luminance responses are similar for the VIVE Pro and Oculus Rift.
Fig. 5. Measured relative luminance as a function of the dark spot diameter in the displayed test patterns for the (a) VIVE Pro and (b) Oculus Rift. The working distance of the probe was fixed at 0.5 mm from the HMD lens for close contact measurements. The diameter of the dark spot was varied up to the size of the bright circular region, which was 50 mm for all cases. The laptop display measurement was used as a reference. Error bars represent standard deviations from four measurements.
For the HMDs, the custom probe with measurement at the top port shows the worst performance overall among the measurement systems despite having a small collimated FOV of around 4 mm (Fig. 5). The relative luminance decreases slowly with an increase in the dark spot size and levels off in the upper 10$^{-2}$ range. The measured signal contamination for large dark spots in the custom probe is more than two order of magnitude higher than the reference measurement. This may be the effect of no baffles inside the integrating sphere for off-axis measurements. As a result, some fraction of unwanted light can always reach the top port, causing the measured values to level off quickly. For small dark spots of less than 5 mm, a commercially available imaging photometer (see Methods for description) showed comparatively better performance among all the measurement systems. Nevertheless, for large dark spots of $\geq$ 5 mm, the imaging photometer still showed high relative luminance in the order of 10$^{-2}$ of the bright region until the 50 mm dark spot completely covered the white region to display a dark image in the HMDs. The contamination of the dark spot signal from the surrounding white region is persistent in our measurements despite using an advanced imaging photometer with low internal glare [32]. These results indicate the presence of significant veiling glare in the VR headsets.
We also investigated the relative luminance response at large working distances of 10 and 20 mm. The measurement probe was moved away from the HMD lens up to 20 mm using a translation stage to be within a typical eye relief distance of VR HMDs [33]. Figure 6(a) presents the relative luminance response of a conventional flat panel laptop display at various working distances for reference. Compared to the close contact measurements at 0.5 mm, the measured relative luminance value per dark spot size correspondingly increases with the working distance of 10 and 20 mm. As the probe moves away from the display surface, more stray light enters the probe and contaminates the dark spot signal. This is in line with the increase in the area of the image visible by the probe with the distance and the unwanted light from the bright regions of the non-Lambertian display reaching the detector. As a result, the relative luminance decreases more slowly with signal contamination, where the required dark spot size to get 1.3 x 10$^{-3}$ background level rises from 8 to 33 mm (Fig. 6(a)). A similar characteristic of the relative luminance response has been reported before while evaluating veiling glare in traditional displays [24,34].
Fig. 6. Measured relative luminance response at three working distances of 0.5, 10, and 20 mm, where (a) is for the laptop display as a reference and (b)-(d) are for the VIVE Pro and Oculus Rift using the various measurement systems.
Figure 6(b)-(d) show no significant change in the relative luminance response with the working distance for the VIVE Pro and Oculus Rift compared to the reference laptop display measurements. The results for the HMDs are consistent regardless of the measurement systems used, whether using the custom probe (Fig. 6(b)) or the commercially available systems (Fig. 6(c) and Fig. 6(d)). These results highlight the difference in the relative luminance response between the conventional flat-panel display and the VR display. VR HMDs have different optical architecture than conventional displays to create a virtual image for a near-eye viewing condition [27,28]. For instance, the VIVE Pro and Oculus Rift include Fresnel lenses between the micro-display panel and the human eyes to create a virtual image at a far distance. In that process, the Fresnel lenses collect the light emitted from the display panel and focus it towards the eyes [35]. The relative luminance within the eye relief distance may not change in the VR HMDs due to this collimating optics providing a large eyebox along the optical axis. Meanwhile, the presence of the same Fresnel lens is likely contributing to high veiling glare in the VIVE Pro and Oculus Rift. A Fresnel lens is well-known to generate stray light and contribute to veiling glare in the light collection process [36–38].
We improved the performance of the custom probe by placing the photodiode at the back port of the integrating sphere. The back port is in the FOV of the probe to directly receive the collimated light (Fig. 1). The back port configuration minimizes the effect of multiple diffuse reflections inside the integrating sphere that may have reduced the probe’s performance during our initial study. Figure 7 shows that the custom probe using the back port configuration performs significantly better than the custom probe using the top port for the relative luminance measurement. The dark background relative luminance drops to 1.4 x 10$^{-3}$ of the bright region similar to the reference measurement values for the 50 mm dark spot size. The custom probe also showed a slightly better performance than the imaging photometer for dark spot size $\geq$ 3.3 mm. In this case, the probe measured about two times lower relative luminance than the imaging photometer until there was no white regions in the displayed test patterns to contaminate the signal. Nevertheless, despite improving the probe’s performance with back port configuration, the overall magnitude of the signal contamination regardless of the dark spot size was still high. Unlike the reference measurement from the laptop display (Fig. 7), the relative luminance never sharply drops to the dark background level in the HMD (VIVE Pro) even when the narrow FOV of the probe is completely covered by the large dark spot (Fig. 4). All these results strongly suggest high level of veiling glare in the HMDs.
Fig. 7. Improving the performance of the custom probe by using the back port of the integrating sphere.
It is worth mentioning that commercially available imaging photometers are becoming increasingly advanced, making them suitable for luminance and glare measurements. Despite some differences in performance, our method may provide certain advantages in specific situations, such as offering a low-cost measurement alternative when top-tier and expensive imaging photometers are not readily accessible to all labs. This allows for a wider range of researchers to engage in HMD luminance characterization, contributing to the advancement of the field.
Although the study was limited to two VR headsets, our method clearly showed the presence of high veiling glare in the consumer-grade HMDs using Fresnel lenses. This high veiling glare issue may make the VR HMDs unsuitable for medical applications that entails grayscale image visualization. Therefore, more accurate and standardized measurement methods are required to evaluate the technical performance of medical VR devices. The main limitation of our current study was including an integrating sphere in the probe design since it did not initially improve the probe’s performance. The integrating sphere allows capturing the image FOV and performing the luminance measurements without disturbing the probe’s alignment by consecutively using its back and top ports, respectively. However, we found that the probe performed poorly for relative luminance measurements using the top port. The spatially-integrated light reaching the top port showed a high relative luminance of >10$^{-2}$ even for a completely dark image with 50 mm dark spot diameter. We saw significant improvement in the probe’s performance by merely using the back port opposite to the light entrance port of the integrating sphere in the FOV of probe. Moreover, in our current setup, the angle of the probe was fixed normal to the HMD display. Future study needs to evaluate the effect of multiple-viewing angles on the relative luminance response by ideally mimicking the motion of the human eye. It is not within the scope of this paper to address the representation of practical eye relief distance through the length of the conic probe. The collimated conic probe using three apertures was designed for small-spot luminance measurements with the goal of minimizing signal contamination from bright regions outside the measurement spot in virtual scenes.
4. Conclusion
We presented an experimental method utilizing a custom conic probe attached to an integrating sphere for characterizing the small-spot luminance of the VR headsets. Despite improving the probe’s performance or using an advanced imaging photometer with low internal glare, we found a significant level of veiling glare in the two consumer-grade VR headsets. We saw that the off-axis ports of integrating sphere are not suitable for luminance measurements. Compared to a conventional flat-panel display, the relative luminance up to 20 mm working distance did not change in the VR HMDs. This was likely due to the unique optical architecture of the HMDs with display light focusing into the eyes for a near-eye viewing condition. Our results overall highlight the issue of high veiling glare in HMDs and the importance of small-spot luminance measurements to adequately evaluate their technical performance for emerging medical applications.
Acknowledgments
ED and NE acknowledge funding by appointments to the Research Participation Program at the Center for Devices and Radiological Health administered by the Oak Ridge Institute for Science and Education through an interagency agreement between the US Department of Energy and the US Food and Drug Administration (FDA). The authors acknowledge support from the White Oak Machine Shop (FDA) for fabricating the conic probe.
Disclosures
All authors declare no conflicts of interest. The mention of commercial products, their resources or their use in connection with material reported herein is not to be construed as either an actual or implied endorsement of such products by the Department of Health and Human Services. This is a contribution of the U.S. Food and Drug Administration and is not subject to copyright.
Data availability
All the necessary data to assess the conclusions can be found in the paper. Additional data related to this work may be obtained from the authors upon reasonable request.
References
1. C. J. Bohil, B. Alicea, and F. A. Biocca, “Virtual reality in neuroscience research and therapy,” Nat. Rev. Neurosci. 12(12), 752–762 (2011). [CrossRef]
2. N. Ahmadpour, H. Randall, H. Choksi, A. Gao, C. Vaughan, and P. Poronnik, “Virtual reality interventions for acute and chronic pain management,” Int. J. Biochem. Cell Biol. 114, 105568 (2019). [CrossRef]
3. L. Wang, M. Huang, R. Yang, H.-N. Liang, J. Han, and Y. Sun, “Survey of movement reproduction in immersive virtual rehabilitation,” IEEE Transactions on Visualization and Computer Graphics 29(4), 2184–2202 (2022). [CrossRef]
4. G. Hattab, A. Hatzipanayioti, A. Klimova, M. Pfeiffer, P. Klausing, M. Breucha, F. v. Bechtolsheim, J. R. Helmert, J. Weitz, S. Pannasch, and S. Speidel, “Investigating the utility of VR for spatial understanding in surgical planning: evaluation of head-mounted to desktop display,” Sci. Rep. 11(1), 13440 (2021). [CrossRef]
5. S. González Izard, J. A. Juanes Méndez, P. Ruisoto Palomera, and F. J. García-Pe nalvo, “Applications of virtual and augmented reality in biomedical imaging,” Journal of medical systems 43, 1–5 (2019). [CrossRef]
6. J. Sutherland, J. Belec, A. Sheikh, L. Chepelev, W. Althobaity, B. J. Chow, D. Mitsouras, A. Christensen, F. J. Rybicki, and D. J. La Russa, “Applying modern virtual and augmented reality technologies to medical images and models,” J. Digit. Imaging 32(1), 38–53 (2019). [CrossRef]
7. E. A. Krupinski and M. Kallergi, “Choosing a radiology workstation: technical and clinical considerations,” Radiology 242(3), 671–682 (2007). [CrossRef]
8. N. Zhu, D. Zhang, W. Wang, X. Li, B. Yang, J. Song, X. Zhao, B. Huang, W. Shi, R. Lu, P. Niu, F. Zhan, X. Ma, D. Wang, W. Xu, G. Wu, G. F. Gao, and W. Tan, “A novel coronavirus from patients with pneumonia in china, 2019,” N. Engl. J. Med. 382(8), 727–733 (2020). [CrossRef]
9. M. Mossa-Basha, J. Medverd, K. F. Linnau, J. B. Lynch, M. H. Wener, G. Kicska, T. Staiger, and D. V. Sahani, “Policies and guidelines for COVID-19 preparedness: experiences from the University of Washington,” Radiology 296(2), E26–E31 (2020). [CrossRef]
10. S. Benbelkacem, A. Oulefki, S. Agaian, T. Trongtirakul, D. Aouam, N. Zenati-Henda, and K. Amara, “Lung infection region quantification, recognition, and virtual reality rendering of CT scan of COVID-19,” Proc. SPIE 11734, 123–132 (2021). [CrossRef]
11. C. Moreira, I. B. Nobre, S. C. Sousa, J. M. Pereira, and J. Jorge, “Improving x-ray diagnostics through eye-tracking and XR,” in 2022 IEEE Conference on Virtual Reality and 3D User Interfaces Abstracts and Workshops (VRW), (IEEE, 2022), pp. 450–453.
12. M. Johnson, C. Zhao, A. Varshney, and R. Beams, “Digital precompensation for luminance nonuniformities in augmented reality head mounted displays,” in 2022 IEEE International Symposium on Mixed and Augmented Reality Adjunct (ISMAR-Adjunct), (IEEE, 2022), pp. 477–482.
13. H. Ha and Y. Kwak, “Prediction model for discomfort luminance levels of head-mounted displays,” Color Res. Appl. 47(4), 1035–1041 (2022). [CrossRef]
14. H. Jung, H.-J. Kim, W.-S. Kang, S. K. Yoo, K. Fujioka, M. Hasegawa, and E. Samei, “Assessment of flat panel LCD primary class display performance based on AAPM TG 18 acceptance protocol,” Med. Phys. 31(7), 2155–2164 (2004). [CrossRef]
15. A. Badano, M. J. Flynn, S. Martin, and J. Kanicki, “Angular dependence of the luminance and contrast in medical monochrome liquid crystal displays,” Med. Phys. 30(10), 2602–2613 (2003). [CrossRef]
16. A. N. Averbukh, D. S. Channin, and M. J. Flynn, “Assessment of a novel, high-resolution, color, AMLCD for diagnostic medical image display: luminance performance and DICOM calibration,” J. Digit. Imaging 16(3), 270–279 (2003). [CrossRef]
17. P. Sund, “Perceived contrast on displays with different luminance ranges,” Med. Phys. 49(4), 2270–2278 (2022). [CrossRef]
18. E.-L. Hsiang, Z. Yang, T. Zhan, J. Zou, H. Akimoto, and S.-T. Wu, “Optimizing the display performance for virtual reality systems,” OSA Continuum 4(12), 3052–3067 (2021). [CrossRef]
19. K. Yin, E.-L. Hsiang, J. Zou, Y. Li, Z. Yang, Q. Yang, P.-C. Lai, C.-L. Lin, and S.-T. Wu, “Advanced liquid crystal devices for augmented reality and virtual reality displays: principles and applications,” Light: Sci. Appl. 11(1), 161 (2022). [CrossRef]
20. E.-L. Hsiang, Z. Yang, Q. Yang, P.-C. Lai, C.-L. Lin, and S.-T. Wu, “AR/VR light engines: perspectives and challenges,” Adv. Opt. Photonics 14(4), 783–861 (2022). [CrossRef]
21. A. Badano, M. J. Flynn, and J. Kanicki, High-fidelity medical imaging displays, vol. 63 (SPIE press, 2004).
22. M. J. Flynn and A. Badano, “Image quality degradation by light scattering in display devices,” J. Digit. Imaging 12(2), 50–59 (1999). [CrossRef]
23. A. Badano, M. J. Flynn, E. Muka, K. D. Compton, and T. L. Monsees, “Veiling glare point-spread function of medical imaging monitors,” Proc. SPIE 3658, 458–467 (1999). [CrossRef]
24. A. Badano and M. J. Flynn, “Method for measuring veiling glare in high-performance display devices,” Appl. Opt. 39(13), 2059–2066 (2000). [CrossRef]
25. K. A. Fetterly, H. R. Blume, M. J. Flynn, and E. Samei, “Introduction to grayscale calibration and related aspects of medical imaging grade liquid crystal displays,” J. Digit. Imaging 21(2), 193–207 (2008). [CrossRef]
26. E. Samei, A. Badano, D. Chakraborty, K. Compton, C. Cornelius, K. Corrigan, M. J. Flynn, B. Hemminger, N. Hangiandreou, J. Johnson, D. M. Moxley-Stevens, W. Pavlicek, H. Roehrig, L. Rutz, E. Samei, J. Shepard, R. A. Uzenoff, J. Wang, and C. E. Willis, “Assessment of display performance for medical imaging systems: executive summary of AAPM TG18 report,” Med. Phys. 32(4), 1205–1225 (2005). [CrossRef]
27. J. Penczek, P. A. Boynton, F. M. Meyer, E. L. Heft, R. L. Austin, T. A. Lianza, L. V. Leibfried, and L. W. Gacy, “Absolute radiometric and photometric measurements of near-eye displays,” J. Soc. Inf. Disp. 25(4), 215–221 (2017). [CrossRef]
28. E. Eisenberg and J. Jensen, “Measuring and qualifying optical performance of AR/VR/MR device displays and addressing the unique visual requirements of transparent AR/MR displays,” Proc. SPIE 11310, 211–229 (2020). [CrossRef]
29. E. Dahal, N. Eby, P. Lemaillet, and A. Badano, “Head-mounted display luminance characterization for medical virtual reality,” Proc. SPIE PC1193, PC1193102 (2022). [CrossRef]
30. C. Fedtke, F. Manns, and A. Ho, “The entrance pupil of the human eye: a three-dimensional model as a function of viewing angle,” Opt. Express 18(21), 22364–22376 (2010). [CrossRef]
31. A. B. Watson and J. I. Yellott, “A unified formula for light-adapted pupil size,” J. Vis. 12(10), 12 (2012). [CrossRef]
32. U. Krüger, I. Rotscholl, and A. Fong, “Critical considerations for characterizing and standardizing ILMDs,” Inf. Disp. 36(3), 25–29 (2020). [CrossRef]
33. Y. Sauer, A. Sipatchin, S. Wahl, and M. García García, “Assessment of consumer VR-headsets’ objective and subjective field of view (FOV) and its feasibility for visual field testing,” Virtual Reality 26(3), 1089–1101 (2022). [CrossRef]
34. M. Choi, J. Wang, W.-C. Cheng, G. Ramponi, L. Albani, and A. Badano, “Effect of veiling glare on detectability in high-dynamic-range medical images,” J. Disp. Technol. 10(5), 420–428 (2014). [CrossRef]
35. K. Bang, Y. Jo, M. Chae, and B. Lee, “Lenslet VR: thin, flat and wide-FOV virtual reality display using fresnel lens and lenslet array,” IEEE Transactions on Visualization & Computer Graphics 27(5), 2545–2554 (2021). [CrossRef]
36. D. J. Lamb and L. W. Hillman, “Computer modeling and analysis of veiling glare and stray light in fresnel lens optical system,” Proc. SPIE 3779, 344–352 (1999). [CrossRef]
37. Y. Geng, J. Gollier, B. Wheelwright, F. Peng, Y. Sulai, B. Lewis, N. Chan, W. S. T. Lam, A. Fix, D. Lanman, Y. Fu, A. Sohn, B. Bryars, N. Cardenas, Y. Yoon, and S. McEldowney, “Viewing optics for immersive near-eye displays: pupil swim/size and weight/stray light,” Proc. SPIE 10676, 19–35 (2018). [CrossRef]
38. J. Xiong, E.-L. Hsiang, Z. He, T. Zhan, and S.-T. Wu, “Augmented reality and virtual reality displays: emerging technologies and future perspectives,” Light: Sci. Appl. 10(1), 216 (2021). [CrossRef]
