A novel method for image clustering

Image clustering has been attracting mounting focus on widely used fields, such as data compression, information retrieval, character recognition and so on, due to the emerging applications of various web-based and mobile-based image retrieval and services. To study this, based on Voronoi diagram, we propose a novel image clustering algorithm to effective discovery of image clusters in this paper. More specifically, based on Voronoi diagrams at first, a number of irregular grids are built across the whole plane. Furthermore, leveraging the good property of “the nearest neighbor” for the Voronoi diagrams, various irregular grids of plane are assigned by the points to different clusters. On the one hand, based on the density of grid points, it automatically adjusts the final suitable number of clustering; on the other hand, according to the changes of the centroids, it tunes the positions for the Voronoi´s seeds. At last, the Voronoi cells finally become the result of clustering process. The empirical experiment results show that our proposed method not only can cluster image dataset effectively, but also can achieve the comparative performance with X-means algorithm and K-means algorithm. Moreover, our proposed method can outperform the effectiveness for both DBSCAN and OPTICS algorithms, which are classic density-based clustering algorithms towards larger-scale real-world applications.