Defocus blur estimation using a Cellular Neural Network

Abstract

Blur identification is one of the most important parts of digital image processing. A conventional blur that occurs in images is out of focus, which is generated because of lens defocus. In this sense, many researchers have presented methods to estimate defocus blur so far. This paper shows how the Cellular Neural Network (CNN) can be used to estimate defocus blur parameter. After the point spread function (PSF) parameter is obtained by the CNN, defocused blur images can be restored effectively.