Blind De-Blurring of Microscopy Images for Cornea Cell Counting

Abstract

Cornea cell count is an important diagnostic tool commonly used by practitioners to assess the health of a patient's cornea. Unfortunately, clinical specular microscopy requires the acquisition of a large number of images at different focus depths because the curved shape of the cornea makes it impossible to acquire a single all-in-focus image. This paper describes two methods and their implementations to reduce the number of images required to run a cell-counting algorithm, thus shortening the duration of the examination and increasing the patient's comfort. The basic idea is to apply de-blurring techniques on the raw images to reconstruct the out-of-focus areas and expand the sharp regions of the image. Our approach is based on blind-deconvolution reconstruction that performs a depth-from-deblur so to either model Gaussian kernel or to fit kernels from an ad hoc lookup table.