Estimation of the Microscope Point Spread Function by Using an Adaptive Image Deblurring Method

Abstract

As one of the most widespread techniques in biological investigation and dynamic process, light compound microscopy has used to analyze the optical properties of biological images. The optical microscope is often used to study and analyze the structures of biological cells and tissues. Microscope images are characterized by a number of specific parameters. Blurring is a major cause of image degradation in optical microscopes. We present an adaptive deconvolution based on the Richardson–Lucy (RL) algorithm to restore micrographs corrupted by out-of-focus blur with parametric estimation of the point spread function (PSF) of the acquisition system by using the blind deconvolution algorithm. This technique improves the quality of digital micrographs with defocus blur effectively. The evaluation algorithm is implemented to improve image quality by removing out-of-focus blur. The algorithm proceeds by randomly generating the PSF at each generation by using non-reference image quality. The PSF is used to estimate the actual image by applying the adaptive RL algorithm. Experimental results show that the proposed blind deconvolution method can estimate the core of the PSF, make good-quality restorted images with a slight ringing effect even, when the image is hardly blurred, and select input images for the non-reference image quality method for deconvolution.

Keywords: Image Processing, Image restoration, Point Spread Function, Deblurring, blind deconvolution.