Fast person-specific image retrieval using a simple and efficient clustering method

Nowadays, more and more interest has been shown in the automatic categorization methods to organize media data as there are increasing number of videos data in people daily life. In the image processing domain, clustering method is the backbone of person specific searching and image retrieval from an image sequence or database. This paper presents a fast image clustering algorithm based on human face, which can be used in an identification biometrics or a face classification system for robot vision. Clustering in different image streams has been achieved in unsupervised manner where no prior knowledge about the input sequence is required. For face detection, both the single Gaussian model and skin color segmentation strategies have been employed. In order to evaluate the performance of the proposed method, 7 different image sequences were used. Experimental results demonstrate that the performance of the proposed method with respect to precision and recall rate is quite satisfactory. Compared with existing methods, the proposed algorithm is more simple and effective.