Head mounted displays (HMD) showed huge market potential in recent years. In these techniques, vergenceaccommodation conflict (VAC) is a fundamental problem which makes viewers feel discomfort and fatigue. To overcome this limitation, researchers proposed many solutions including Maxwellian view displays, vari-focal plane displays, multifocal plane displays, integral imaging-based displays, and computational multilayer displays. These techniques can enable correct or nearly correct focus cues, however, they failed to achieve both high image fresh rate and high lateral resolution with a compact architecture. In this paper, we propose a compact birefringent-based virtual reality (BVR) HMD with correct focus cues by spatially projecting the input images onto four depth planes. In the BVR, two stacked liquid crystal displays (LCDs) provide two axially separated input images in an additive fashion. We set a liquid crystal panel behind the LCDs to modulate the polarization of the emitting light from the LCDs pixel-wise. After that, a birefringent lens and an eyepiece project the modulated light onto four depth planes at 0D, 1D, 2D, and 3D. To minimize the astigmatism of the system, we employ a birefringent doublet with orthogonal optic axes and use an eyepiece to suppress the overall aberration. Comparing to the existing techniques, the proposed BVR mitigates the VAC problem with a compact architecture. Moreover, because there is no temporal multiplexing and lateral resolution sacrifice, the BVR can easily achieve high image refresh rate and high lateral resolution. Herein, we present the optical design of the BVR and characterize its performance in Zemax.
Human vocal fold vibration is a complex 3D movement, and its frequency varies from 85-150 Hz (typical males) to 165-250 Hz (typical females). The unusual 3D shape of vocal folds is a hallmark for a variety of vocal diseases, such as polyps, nodules, recurrent nerve paralysis, and cancer.
The standard in-office methods for diagnosing voice disorders encompass videostroboscopy and high-speed videoendoscopy. Both techniques image only the horizontal movement of vocal folds. They cannot measure the absolute vibrational movement of vocal folds along the air flow direction. Despite the vital importance, currently, very few methods are available for 3D laryngeal imaging.
To overcome above limitation, we introduce the paradigm of light field imaging into laryngoscopy. The resultant method, which we term light field laryngoscopy (LFL), will enable 3D imaging of vocal folds in a single camera exposure. Moreover, to alleviate the trade-off between the spatial- and depth-resolution in light field imaging, we developed a hybrid imaging scheme which comprises an additional camera to provide a high-resolution 2D reference. Herein we will present the optical design of LFL, and characterize the imaging performance of the prototype.
The convergence of recent advances in optical fabrication and digital processing yields a generation of imaging technology—light-field (LF) cameras which bridge the realms of applied mathematics, optics, and high-performance computing. Herein for the first time, we introduce the paradigm of LF imaging into laryngoscopy. The resultant probe can image the three-dimensional shape of vocal folds within a single camera exposure. Furthermore, to improve the spatial resolution, we developed an image fusion algorithm, providing a simple solution to a long-standing problem in LF imaging.
A fast and precise registration method for multi-image snapshot Fourier transform imaging spectroscopy is proposed. This method accomplishes registration of an image array using the positional relationship between homologous points in the subimages, which are obtained offline by preregistration. Through the preregistration process, the registration problem is converted to the problem of using a registration matrix to interpolate subimages. Therefore, the hardware interpolation of graphics processing unit (GPU) texture memory, which has speed advantages for its parallel computing, can be used to significantly enhance computational efficiency. Compared to a central processing unit, GPU performance showed ∼27 times acceleration in registration efficiency.
Fourier-transform imaging spectrometers are rapidly developed due to their extensive use in industrial monitoring, target detection, and chemical identification. Static Fourier-transform imaging spectrometer (SFIS) containing a birefringent interferometer is one of the most popular directions due to its inherent robustness. However, the SFIS suffers from its low achievable signal-to-noise ratio (SNR) because of the restriction of incident angle. Meanwhile, in applications, the SNR is perhaps the most important factor to determine the usefulness of an instrument. In this paper, we report here a Static Fourier-transform imaging spectrometer based on differential structure (SFIS-DS) in the 400-800nm wavelength range with a high SNR. As in electronic system, the differential structure can double optical efficiency and strongly restrain common mode error in the SFIS-DS. And the differential structure described here is also available for any instruments containing a birefringent interferometer. However, the drawback of the SFIS-DS is that the two images obtained by the two differential channels need precise registration which can be overcome by a sub-pixel spatial registration algorithm. The experimental results indicate the SFIS-DS can increase the SNR by no less than 40%.