Graph Object Oriented Database for Semantic Image Retrieval

This paper presents a new method for image retrieval using a graph object oriented database for processing the information extracted from the image through the segmentation process and through the semantic interpretation of this information. The object oriented database schema is structured as a classes hierarchy based on graph data structure. A graph structure is used in all phases of the image processing: image segmentation, image annotation, image indexing and image retrieval. The experiments showed that the retrieval can be conducted with good results and the method has a good time complexity.