A New Parallel Hierarchical K-Means Clustering Algorithm for Video Retrieval

The K-means clustering algorithm has been widely adopted to build vocabulary in image retrieval. But, the speed and accuracy of K-means still need to be improved. In the manuscript, we propose a New Parallel Hierarchical K-means Clustering (PHKM) Algorithm for Video Retrieval. The PHKM algorithm improves on the K-means as the following ways. First, the Hellinger kernel is used to replace the Euclidean kernel, which improves the accuracy. Second, the multi-core processors based parallel clustering algorithm is proposed. The experiment results show that the proposed PHKM algorithm is very faster and effective than K-means.