An Approach for Image Analysis Using Complex Network

Abstract

The information of an image is characterized not only by the intensity of individual pixels, but also by the correlation of intensity among pixels in the surrounding area. Recently, the complex network has become an interest topic of research in many fields including the image processing. This paper presents an approach of image analysis using the complex network. In particular, a complex network is constructed and analyzed using the data of images, i.e. entropy, intensity, and position (EIP). In the EIP method, the input image is divided into equal-sized square windows; each window is considered as a node of the complex network. Links among adjacent nodes are determined based on comparing the difference in average intensity and entropy of neighboring windows with some thresholds. The created network is analyzed using common measures of complex network.