Options
1992
Conference Paper
Title
Object-oriented image database for an open imaging environment
Abstract
apART, a system for the acquisition, processing, archiving, and retrieval of digital images has been developed. The platform of the system is a general-purpose workstation which allows the processing and interactive high-resolution raster images in cooperation with otheration with other heterogeneous components like acquisition or archiving devices. To further this end, the workstation provides open image communications capabilities (via either a local in-house network or a telecomork) and image input from such devices as CCD-cameras as CCD-cameras and slide scanners. In this way the system provides an open, distributed imaging environment. The system is driven by an object-oriented user interface. Devices (image sources and destinations), operators (derived from a commercial image processing library), and images (of different data types) are managed and presented uniformly to the user. An essential part of the system is the image databases, which support network access to both images and image-related data. An image database is built on top of the structurally object-oriented database system APRIL and is outlined in the sequel. A digital image can be classified within the database as one image object, which is composed of the following components: the original image in maximum quality and resolution, several derivatives for the graphical-interactive presentation and/or retrieval, and image related attributes (e.g. resolution, channels, colormap). Each image object is an instance of an user defined image type which describes, besides the above mentioned components, specific semantic attributes of the attached image objects. The image types of one image database form a directed, acyclic classification graph which can be extended by using inheritance mechanisms. A new image type inherits all image attributes of its direct and indirect predecessors within the graph and can be extended with additional user defined attributes. Thus, the image database can be structured
Conference