IMAGE-ANALYSIS - PROBLEMS, PROGRESS AND PROSPECTS

Much of the work on applications of pattern recognition, and a significant fraction of the work in artificial intelligence, has dealt with the analysis and interpretation of images. This subject has been variously known as pictorial pattern recognition, image analysis, scene analysis, image understanding and computer vision. Its applications include document processing (character recognition, etc. ), microscopy, radiology, industrial automation (inspection, robot vision), remote sensing, navigation and reconnaissance, to name only the major areas. Basic steps are outlined in a general image analysis process. The state of the art is summarized with respect to each step, points out limitations of present methods and indicates potential directions of future work. Feature extraction, texture analysis, surface estimation, orientation, image matching, range estimation, segmentation, object representation and model matching are the state of the art with respect to each step, points out limitations of present methods and indicates potential directions for future work.