Shape-Based Image Classification And Retrieval

Shape descriptors have been used frequently to characterize an image for classification and retrieval tasks. For example, shape features are applied in medical imaging to classify whether an image shows a tumor or not. The patent office uses the similarity of shape to ensure that there are no infringements of copyrighted trademarks. This paper focuses on using machine learning and information retrieval techniques to classify an image as belonging to one of many classes based on its shape. In particular, we compare support vector machines, Naive Bayes, maximum entropy, linear discriminant analysis, and relevance-based language models for classification. Our results indicate that, on the standard MPEG-7 database, the relevance model outperforms the machine learning techniques and is competitive with prior work on shape-based retrieval with a classification rate of 79.8%. We also show that the relevance model can be easily extended to classify images with multiple labels. The relevance model approach may be used to perform shape retrieval using keywords. Experiments on the MPEG-7 database and a binary version of the COIL-100 database show that good retrieval performance may be obtained using a small number of examples even when there is a large amount of viewpoint change.