Image Extrema Analysis and Blur Detection with Identification

In real image processing applications, images may be blurred or not. When blur is present, the type and degree of degradation vary from one image to another. The process of restoring these images are usually computationally demanding so that there is a need to first detect blurs. If an image is not blurred then it need not undergo the restoration process. In this work, a novel algorithm that simultaneously detects and identifies blurs, is proposed. This method is based on the analysis of extrema values in an image. The extrema histograms are first constructed then analyzed in order to extract feature values. The distinctness of these values in the presence of blur is used. It is computationally simple and fast thereby making it suitable for preprocessing especially in practical imaging applications. Experimental results on natural images and its synthetically blurred versions show the validity of the proposed method.