Invariants based blur classification algorithm

Abstract

Extraction of information from an image acquired by real imaging systems is a difficult task, since the observed image may be degraded by blurring. In this paper, a framework for classification of blur in an image is presented and a technique for classification of blur using invariants is proposed. In this method, the blur classification is carried out without estimating the blurring function. The proposed technique is applied on a large dataset of images degraded by motion blur, Gaussian blur and defocus blur. The simulation results show that the proposed method gives accurate classification of the blur present in an image.