3. A New Hierarchical Approach for Image Clustering

The key problem in achieving efficient and user-friendly retrieval in the domain of image is the development of a search mechanism to guarantee delivery of minimal irrelevant information (high precision) while ensuring that relevant information is not overlooked (high recall). The unstructured format of images tends to resist the deployment of standard search mechanism and classification techniques. As a method to provide better organization of images, clustering is an important aspect for effective image retrieval. In this chapter, we introduce a clustering method based on unsupervised neural nets and self-organizing maps. In dynamic growing self-organizing tree algorithm (DGSOT), a hierarchy is constructed from top to bottom. We observe that DGSOT outperforms the traditional Hierarchical Agglomerative Clustering (HAC) algorithm in terms of E-measure.