Blind Evaluation of Image Scrambling Degree based on the Correlation of Adjacent Pixels

Abstract

The correlation of adjacent pixels is one of the important parameters for measuring the distribution of adjacent pixels’ gray value. And it is widely used to evaluate the scrambling degree of digital image. Distribution figures are always plotted to show the relation of adjacent pixels’ gray value, but only subjective and qualitative analysis has been applied to explain the uniformity of dots in these figures so far. In order to evaluate their uniformity quantitatively and objectively, two approaches are advanced: average gray difference method and block uniformity method. And, two blind evaluating methods on the image scrambling degree are presented further according to the former. Simulation results show that both of the two approaches can describe the uniformity of distribution figures of adjacent pixels’ gray value quantitatively and objectively. They can be operated with just a little data of a cipher image and without using any data of plain image. These two evaluation methods on image scrambling degree are in a good agreement with the Human Vision System, and they can achieve the goal of objective and fast blind evaluation. DOI: http://dx.doi.org/10.11591/telkomnika.v11i11.3496