×

A hierarchical image annotation method based on SVM and semi-supervised EM. (English) Zbl 1240.68175

Summary: Automatic image annotation, which aims at automatically identifying meaningful objects in a digital image and then assigning semantic keywords to them, is not a very difficult task for humans but is regarded as a difficult and challenging problem for machines. In this paper, we present a hierarchical annotation scheme based on the observation that, in general, human visual identification of a scenery object is a rough-to-fine hierarchical process. First, the input image is segmented into multiple regions and each segmented region is roughly labeled with a general keyword using a multi-classification support vector machine. Since the results of rough annotation affect fine annotation directly, we construct a statistical contextual relationship to revise the improper labels and improve the accuracy of the rough annotation. To obtain a reasonably fine annotation for the roughly classified regions, we propose an active semi-supervised expectation-maximization algorithm, which can not only find the representative pattern of each fine class but also classify the roughly labeled regions into corresponding fine classes. Finally, the contextual relationship is applied again to revise the improper fine labels. To illustrate the effectiveness of the presented approaches, a prototype image annotation system is developed, the preliminary results of which show that the hierarchical annotation scheme is effective.

MSC:

68T05 Learning and adaptive systems in artificial intelligence
68T10 Pattern recognition, speech recognition
68U10 Computing methodologies for image processing
PDFBibTeX XMLCite
Full Text: DOI