Yibin Tian

Performance of focus measures in the presence of nondefocus aberrations

The purposes of the study were to compare the performance of ten representative focus measures in the presence of nondefocus aberrations and to evaluate their applicability to the eye. For fixed amounts of nondefocus aberrations, the amount of defocus was changed to generate a series of blurred images from which focus measure curves were derived. In the presence of small amounts of nondefocus aberrations, all focus measures showed unimodal and monotonic behavior, although there were large differences in their sensitivity to defocus and effective ranges. There were breakdowns in monotonicity and unimodality for some focus measures when applied to data from human eyes, while other focus measures could detect the shift in the best-focus plane in the blurred image series resulting from spherical aberration.

The significance of retinal image contrast and spatial frequency composition for eye growth modulation in young chicks.

PURPOSE This study sought further insight into the stimulus dependence of form deprivation myopia, a common response to retinal image degradation in young animals. METHODS Each of 4 Bangerter diffusing filters (0.6, 0.1, <0.1, and LP (light perception only)) combined with clear plano lenses, as well as plano lenses alone, were fitted monocularly to 4-day-old chicks. Axial ocular dimensions and refractive errors were monitored over a 14-day treatment period, using high frequency A-scan ultrasonography and an autorefractor, respectively. RESULTS Only the <0.1 and LP filters induced significant form deprivation myopia; these filters induced similarly large myopic shifts in refractive error (mean interocular differences+/-SEM: -9.92+/-1.99, -7.26+/-1.60 D, respectively), coupled to significant increases in both vitreous chamber depths and optical axial lengths (p<0.001). The other 3 groups showed comparable, small changes in their ocular dimensions (p>0.05), and only small myopic shifts in refraction (<3.00 D). The myopia-inducing filters eliminated mid-and-high spatial frequency information. CONCLUSIONS Our results are consistent with emmetropization being tuned to mid-spatial frequencies. They also imply that form deprivation is not a graded phenomenon.

Evaluation of retinal image degradation by higher-order aberrations and light scatter in chick eyes after photorefractive keratectomy (PRK)

Intraocular light scatter can severely degrade retinal image quality, a key determinant of visual performance and eye growth regulation, yet it is often ignored in analyses of retinal image quality. A method for characterizing ocular light scatter using Shack–Hartmann images is described here. A local light scatter map is derived by de-convolving the dots in the Shack–Hartmann image recorded from a real eye with those from a model eye, and used to estimate overall ocular light scatter. As an index of retinal image quality, the modulation transfer function (MTF) of the eye is derived as the product of MTFs for light scatter and aberrations alone and for comparative purposes, an MTF volume ratio, equivalent to the Strehl ratio computed in the frequency domain, is computed. These analyses were applied to Shack–Hartmann images collected from three chick eyes, one day before and four days after photorefractive keratectomy (PRK) surgery. PRK was performed on one eye of each of three 4-week-old chicks; ablation parameters for a clinical argon fluorine excimer laser system from Visx® were adjusted based on the results of ablations performed on enucleated chick eyes of the same age. While all three eyes showed increased higher order aberrations after PRK, only two eyes showed significantly increased light scatter. The proposed method for characterizing retinal image quality was able to capture these interocular differences.

Do focus measures apply to retinal images

The diverse needs for digital auto-focusing systems have driven the development of a variety of focus measures. The purpose of the current study was to investigate whether any of these focus measures are biologically plausible; specifically whether they are applicable to retinal images from which defocus information is extracted in the operation of accommodation and emmetropization, two ocular auto-focusing mechanisms. Ten representative focus measures were chosen for analysis, 6 in the spatial domain and 4 transform-based. Their performance was examined for combinations of non-defocus aberrations and positive and negative defocus. For each combination, a wavefront was reconstructed, the corresponding point spread function (PSF) computed using Fast Fourier Transform (FFT), and then the blurred image obtained as the convolution of the PSF and a perfect image. For each blurred image, a focus measure curve was derived for each focus measure. Aberration data were either collected from 22 real eyes or randomly generated data based on Gaussian parameters describing data from a published large scale human study (n>100). For the latter data set, analyses made use of distributed computing on a small inhomogeneous computer cluster. In the presence of small amounts of nondefocus aberrations, all focus measures showed monotonic changes with positive or negative defocus, and their curves generally remained unimodal, although there were large differences in their variability, sensitivity to defocus and effective ranges. However, the performance of a number of these focus measures became unacceptable when nondefocus aberrations exceed a certain level.

Retinal image degradation by optical aberrations and light scatter in normal and albino chick eyes

Comprehensive evaluation of retinal image quality requires that light scatter as well as optical aberrations be considered. In investigating how retinal image degradation affects eye growth in the chick model of myopia, we developed a simple method based on Shack-Hartmann images for evaluating the effects of both monochromatic aberrations and light scatter on retinal image quality. We further evaluated our method in the current study by applying it to data collected from both normal chick eyes and albino eyes that were expected to show increased intraocular light scatter. To analyze light scatter in our method, each Shack-Hartmann dot is treated as a local point spread function (PSF) that is the convolution of a local scatter PSF and a lenslet diffraction PSF. The local scatter PSF is obtained by de-convolution, and is fitted with a circularly symmetric Gaussian function using nonlinear regressions. A whole-eye scatter PSF also can be derived from the local scatter PSFs for the analyzed pupil. Aberrations are analyzed using OSA standard Zernike polynomials, and aberration-related PSF calculated from reconstructed wavefront using fast Fourier transform. Modulation transfer functions (MTFs) are computed separately for aberration and scatter PSFs, and a whole-eye MTF is derived as the product of the two. This method was applied to 4 normal and 4 albino eyes. Compared to normal eyes, albino eyes were more aberrated and showed greater light scatter. As a result, overall retinal image degradation was much greater in albino eyes than in normal eyes, with the relative contribution to retinal image degradation of light scatter compared to aberrations also being greater for albino eyes.

Monte Carlo Evaluations of Ten Focus Measures

A high-performance focus measure is one of the key components in any autofocus system based on digital image processing. More than a dozen of focus measures have been proposed and evaluated in the literature, yet there have be no comprehensive evaluations that include most of them. The purpose of the current study is to evaluate and compare the performance of ten focus measures using Monte Carlo simulations, run on a self-built scalable inhomogeneous computer cluster with distributed computing capacity. From the perspective of a general framework for focus measure evaluations, we calculate the true point spread functions (PSFs) from aberrations represented by OSA standard Zernike polynomials using fast Fourier transform. For each run, a range of defocus levels are generated, the PSF for each defocus level is convoluted with an original image, and a certain amount of noise is added to the resulting defocused image. Each focus measure is applied to all the blurred images to obtain a focus measure curve. The procedure is repeated on a few representative images for different types and levels of noise (Gaussian, salt & pepper, and speckle). The performance of the ten focus measures is compared in terms of monotonicity, unimodality, defocus sensitivity, noise sensitivity, effective range, computational efficiency and variability.