Optimization Of Zonal Wavefront Estimation And Curvature Measurements

摘要: Optical testing in adverse environments, ophthalmology and applications where characterization by curvature is leveraged all have a common goal: accurately estimate wavefront shape. This dissertation investigates sensing techniques as applied to optical based on gradient measurements. Wavefront involves the ability shape over any aperture geometry, which requires establishing sampling grid estimation scheme, quantifying errors caused measurement noise propagation, designing an instrument with sufficient accuracy sensitivity for application. Starting gradient-based sensing, zonal least-squares algorithm irregular pupil size presented, normal matrix equation sets share pre-defined matrix. A Gerchberg–Saxton iterative method employed reduce deviation estimated across discontinuous boundary. The results show that RMS error of from original can be less than λ/130~ λ/150 (for λ equals 632.8nm) after about twelve iterations λ/100 few four iterations. presented approach handling shapes applies equally well data. defining characteristic its propagation behavior. coefficient formulated function eigenvalues estimation-related matrices, such functions are established each basic geometries (i.e. Fried, Hudgin Southwell) serial numbering ii square array sequentially indexed row row. piston-value fixed, odd-number sizes yield lower even-number geometries. Fried geometry either allows sub-sized estimations within domain or yields two-rank deficient full aperture; but latter usually suffers high waffle mode problem. offers propagator between those Southwell For both difference-based estimations, shown offer lowest minimum-norm solution. Noll’s theoretical result, was extensively used reference previous literature estimate, corresponds size. curvature-based concept differential Shack-Hartmann (DSH) sensor proposed. derived collimated beam split into three output channels, along lenslet located. Three Hartmann arrays generated arrays. Two lenslets shear two perpendicular directions relative third one. By quantitatively comparing coordinates differentials slope at point obtained, so Laplacian curvatures twist terms will available. acquisition using Hartmann-based us uniquely determine principal more prior methods. Measurement local opposed slopes unique because intrinsic under test, it absolute measurement. least-squares-based developed iii data, validated. An implementation DSH proposed experimental system this initiated. shares important features Roddier’s sensor. It two-dimensional parallel Because sensor, provides measurements thus insensitive vibrations, tip/tilts, whole body movements. does not suffer other sources errors, scanning noise. Combined algorithm, low mid frequencies may recovered. Notice operates therefore difficulty associated operation close caustic zone avoided. Finally, DSH-curvature-sensor-based 2π-ambiguity problem, potentially small large aberrations measured.