Ophthalmoscopes to image the retina are widely used diagnostic tools in ophthalmology and are vital for the early detection of many eye diseases. Although there are various effective optical implementations of ophthalmoscopes, new, robust systems may have a future practical importance in cases where ocular media present significant opacities. Here, we present, as a proof of concept, a novel approach for imaging the retina in real time using a single pixel detector combined with spatially coded illumination. Examples of retinal images in both artificial and real human eyes are presented for the first time to our knowledge.

Single pixel imaging can be the preferred method over traditional 2D-array imaging in spectral ranges where conventional cameras are not available. However, when it comes to real-time video imaging, single pixel imaging cannot compete with the framerates of conventional cameras, especially when high-resolution images are desired. Here we evaluate the performance of an imaging approach using two detectors simultaneously. First, we present theoretical results on how low SNR affects final image quality followed by experimentally determined results. Obtained video framerates were doubled compared to state of the art systems, resulting in a framerate from 22 Hz for a $32\times 32$ resolution to 0.75 Hz for a $128\times 128$ resolution image. Additionally, the two detector imaging technique enables the acquisition of images with a resolution of $256\times 256$ in less than 3 s.

Purpose: We report on a semi-automated system for frequency analysis of the Stiles–Crawford effect of the first kind (SCE-I) using flicker methodology designed to gain insight into the temporal dynamics of the perceived visibility for alternating pupil entrance points. We describe the system and its calibration in detail and discuss psychophysical measurement data obtained for the two authors. Methods: A uniaxial system is used for SCE-I characterization of two emmetropic subjects as a function of flicker frequency for narrow wavelength bands chosen in the range of 450–700 nm using a fibre-guided tungsten–halogen lamp as light source. The flicker is realized using two orthogonally mounted galvanometric scanning mirrors that allow linear trajectories at any angle across the pupil. A fast tuneable liquid-crystal neutral density filter is used for brightness adjustment and another liquid-crystal filter is used for wavelength adjustment at each pupil point allowing simultaneous hue-shift determination for the Stiles–Crawford effect of the second kind (SCE-II). Results: Validation of the system is realized with a CCD camera, a spectrometer and a powermeter, and the data obtained are used in the software to calibrate all subsequent human subject measurements. The psychophysical data obtained show a strong frequency dependence of the Gaussian SCE-I with a characteristic directionality parameter, ρ, that is found to increase from 0.03 to 0.06/mm² with flicker in the range of 1–10 Hz. The simultaneously determined hue shift could not be determined beyond 1 Hz due to the longer time required for a subjective determination. Conclusion: We have reported on a fast uniaxial system for temporal characterization of the SCE-I. The psychophysical results obtained show that accurate specification of frequency in flicker analysis is mandatory when comparing SCE-I visibility and directionality curves obtained with those obtained using quasi-static bipartite fields. A uniaxial design offers unique advantages over that of common two-channel systems by completely eliminating spectral errors or brightness differences in the two branches that otherwise will impose on those of the visual system and degrade the psychophysical data. Future work with more subjects will be used to narrow the uncertainty and the causes of the effects observed.