Medical imaging is shifting from film to electronic images. The STARE (structured analysis of the retina) system is a sophisticated image management system that will automatically diagnose images, compare images, measure key features in images, annotate image contents, and search for images similar in content. We concentrate on automated diagnosis. The images are annotated by segmentation of objects of interest, classification of the extracted objects, and reasoning about the image contents. The inferencing is accomplished with Bayesian networks that learn from image examples of each disease. This effort at image understanding in fundus images anticipates the future use of medical images. As these capabilities mature, we expect that ophthalmologists and physicians in other fields that rely in images will use a system like STARE to reduce repetitive work, to provide assistance to physicians in difficult diagnoses or with unfamiliar diseases, and to manage images in large image databases. 1.

This paper describes steps towards an information system for the storage and content-based retrieval of ocular fundus images. Based on the Virage Incorporated framework for defining similarity metrics, the authors have developed a number of primitives for the representation of ocular fundus images. A prototype Query By Pictorial Example (QBPE) system yields similarity rankings in approximate agreement with those of a human expert.

No abstract available.

No abstract available.

The authors show how stereo-based range data, obtained over time by a moving vehicle can be integrated without the explicit knowledge or computation of camera motion. A unique aspect of the approach is the transformation of range information into relative distances and encoding these distances in image registered maps. Results of experiments on dynamic stereo images, are presented.>

Fusion of 3-D spatial information obtained by a dynamic vision system is addressed. The representation of information is seen to be critical in forming a solution. An object-centered representation that encodes, in image registered maps, the relative distances of objects to a set of scene referents is proposed. This representation facilitates fusion of dynamic spatial information without direct use or knowledge of camera motion and points the way for a new model for dynamic vision.>

Stereo algorithms using a wide baseline must cope with a large search space on top of the problems due to noise. A single pair of images is not a reliable source of depth information. One solution is to integrate the information from several image pairs. By incrementally refining estimates of the depth as the camera changes position, it is possible to build a description of the scene which, in turn, can be used as feedback to improve the extraction of depth information. The authors have developed a stereo algorithm based on correlation matching that is especially suited to being integrated into a proposed feedback loop. They have begun experiments with the integration of depth information over a sequence into a single environmental map.>