Analysis of autofocusing evaluation functions of intraocular lens

Five autofocusing evaluation functions were compared for monofocus and multifocus intraocular lenses. In order to find the region of interests in an images, a cross-correlation coefficient between the template and target image was computed. We used autofocusing evaluation function i.e. regional spectral weights (SPEC), wavelet transform (WAV), variance (VAR), gradient (GRAd), and Laplacian filter (LAP). Experimental results show that LAP sensitively discriminates the characteristics of multifocus IOL. The shape of the evaluation function was further spreaded outward from the LAP, SPEC, GRAd, WAV, and VAR.