THE COMPARATIVE ANALYSIS OF THE OPERATORS USED FOR EDGE DETECTION OF THE IMAGE FOR THE ADJUSTMENT OF SHARPNESS

Abstract

The visual perception of subjects is appreciably defined by sharpness of their borders (edges). The image having fine color balance, but not having a clear boundary between color areas, can be perceived as unsatisfactory on quality. Therefore sharpness adjustment is an important stage of image processing. Adjustment can be carried out by various modes. Before decision-making on a choice of the strategy of adjustment of sharpness it is necessary to carry the image to one of the groups from the point of view of semantics. It can be made on the basis of the analysis of the egdes. For this reason special attention should be given reliability of these procedures. Now for the decision of this problem there is a set of program operators, such as Canni-operator, LoG-operator, Sobel-operator, Prewitt-operator and Roberts-operator. These operators use various algorithms of allocation of important borders on the image. The main goal of this experiment was the comparative analysis of different operators and revealing of most effective of them. The special attention was given also to the analysis of histograms of received images (quantity of pixels which concern edges). It gives some recommendations of practical importance. It has allowed to chotose most effective of operators, which can be used further for the purpose of the adjustmens of sharpness.