Interest Point Detection in Images by a Local Centrality Algorithm on Complex Networks
The field of Digital Image Processing offers an interesting framework for the application of the Complex Networks theory. Since images can be seen as organized data structures of adjacent pixels, it becomes natural to model and analyze them using complex network properties. We present a local centrality algorithm on a network with the aim detect the position and importance of interest points (i.e. corners) in a digital image. An spatial and weighted complex network is associated to the image and a new method for locating these feature points based on a local centrality measure of the corresponding network, is proposed.