Computer vision and image processing research is carried out in both GRAIL (the Graphics and Imaging Laboratory in CSE) and ISL (the Intelligent Systems Laboratory in EE). The following are some current projects:
Graphics and image processing technology can greatly enrich the sensory experiences of students studying mathematics. The project, "Mathematics Experiences Through Image Processing", funded by the National Science Foundation program on Applications of Advanced Technologies within the Directorate for Education and Human Resources, is working to bring exciting learning activities to K-12 students and teachers including those of an inner-city middle school in Seattle. The project currently supports eighth-grade students in manipulating scanned images on personal computers and learning about mathematical representation of images, functions, coordinate systems, and the measurement of shape, as well as other topics in mathematics. Recent technical developments within the project include (1) the "METIP Programming Environment" which permits both rapid prototyping of experimental image-processing courseware under Windows and a new image-oriented approach to teaching programming, and (2) "Multi-In" software and special hardware to support cooperative learning on PCs using multiple input devices.
The goal of the Intelligent Machine Vision project is to develop an automated vision system for inspection and robot guidance that bridges the gap from CAD models to machine vision algorithms. The PREMIO system converts a CAD model to a model that is appropriate for machine vision, uses the vision model to predict features that will appear in images under various lighting and other environmental conditions, and uses the predictions to guide a matching procedure that finds correspondences between image features and model features for estimation of position and orientation. A new project is the development of an active object recognition system for analysis of multi-object scenes. The system will use a single, movable camera and several light sources. It will generate hypotheses about the scene from several images taken from a single viewpoint with different light sources. The hypotheses will then be used to determine an action such as moving the camera or light sources and taking more images with which to verify or disprove the hypotheses. After several iterations of image acquisition, image processing, analysis, actions, the system will produce an explanation of the scene in terms of the models that it finds present and their positions in the scene.
Using the VLSI design facilities in the department, several image processing chips have been designed and fabricated. One of these, the "Systolic Cellular Logic" chip was specially designed for rapid computation of local-neighborhood binary image operations, and it was used as the principal building block in a highly parallel image processor "SCAM" (Systolic Cellular Array Machine). A compiler and visual simulator for this system have been developed.
Finally, a visual database system for computer vision research is being developed. Robotics, remote sensing, and medical applications are being used to motivate the design of this system. The system can store images, sequences of images, and structures extracted from images along with ancillary information. It has an interactive visual front end that allows users, who are expected to be vision researchers, to examine images and other intermediate structures as they test their algorithms. A new related project is to develop a scientific database and experiment management system for NASA researchers that runs on massively parallel machines.
Principal Investigators: Shapiro, Tanimoto, Brinkley (Biological Structure), Haralick (Electrical Engineering), Kalet (Radiation Oncology), and King (Mathematics).