Scale Space Exploration For Mining Image Information Content

Images are highly complex multidimensional signals, with rich and complicated information content. For this reason they are difficult to analyze through a unique automated approach. However, a hierarchical representation is helpful for the understanding of image content. In this paper, we describe an application of a scale-space clustering algorithm (melting) for exploration of image information content. Clustering by melting considers the feature space as a thermodynamical ensemble and groups the data by minimizing the free energy, having the temperature as a scale parameter. We develop clustering by melting for multidimensional data, and propose and demonstrate a solution for the initialization of the algorithm. Due to the curse of dimensionality, for initialization of clusters we choose the initial clusters centers with an algorithm that performs a fast cluster centers estimation with low computation cost. We further analyze the information extracted by melting and propose a structu...

• Type: Conference
• Year: 2002
• Where: KDD
• Authors: Mariana Ciucu, Patrick Héas, Mihai Datcu, James C. Tilton
• Keywords: Compact Data Structure, Data Cluster Structure, Data Mining, KDD 2002, Scale-space Clustering Algorithm