Abstract
Purpose: :
The correct ocular modulation transfer function (MTF) in the full spatial frequency range and the ocular phase transfer function (PTF) in the low spatial frequency range can be obtained with the double–pass point–spread function (PSF) measurement system using different aperture sizes. However, for reconstructing the correct single–pass PSF, the ocular PTF in the high spatial–frequency range is required. We propose a new phase–retrieval algorithm to replace the iterative methods in space and the Fourier domain.
Methods: :
Computer simulation shows that the real and imaginary parts of the ocular optical transfer function (OTF) have changed gradually in a wave–like fashion. Therefore, the values of the real and imaginary parts of the ocular OTF can be extrapolated from the low frequency to the high spatial frequency value using the definition that the sum of the second intensity of a real part and that of an imaginary part of the OTF is equal to that of MTF as the limited condition. This technique was tested with many PSFs deduced from the wavefronts by changing the Zernike coefficient to the 4th order by computer and also with double–pass data obtained using Topcon prototype equipment.
Results: :
The degree of the recovery was estimated by the similarity of the PSF using the correlation between the original PSF and each PSF, which was obtained before and after phase recovery. The mean correlation coefficient was 0.1 before phase recovery and 0.38 after phase recovery. The expected PSFs were reconstructed from double–pass data obtained by Topcon prototype equipment, indicating that this algorithm is effective.
Conclusions: :
We proposed a new phase–retrieval algorithm with the correct ocular MTF and the limited ocular PTF. Moreover, the validity of this technique was verified and this algorithm will likely be effective for clinical application.
Keywords: optical properties • image processing