Fast hierarchical image segmentation through irregular pyramids

Scene Structural matrices (SSMs) are 2-D tables that can be used conveniently to capture the organization of a scene in a color image when it has been adaptively partitioned and represented by a bintree. In this paper, we present a number of methods to populate the cells of SSMs. Well-established and popular image indexing features are used. Because the SSMs implicitly contain scene organization information and intermediate level concept, indexing the cells with these features is equivalent to adapting them to the structure of the scene. We present experimental results on retrieving one type of images which can be described by an intermediate level concept -- the landscape images. Results showed that the new methods significantly outperformed traditional methods.