Interactive map conversion: combining machine vision and human input

Abstract

The authors present an interactive map conversion system which combines a human operator's high level reasoning with machine perception under the Human-Machine Perceptual Cooperation (HMPC) paradigm. HMPC defines two channels of interaction: the focus of attention (FOA) by which the user directs the attention of machine perception, and context. As the user moves the FOA across a raster map display via a pointing device, a smart cursor operates proactively on the data highlighting objects for extraction. The FOA permits foveal emphasis, enabling the user to vary motor precision with map clutter. HMPC provides for contexts at four levels of abstraction. This permits the efficiency of the system to degrade gracefully as data quality worsens. They also present a boundary-based line follower which computes line thickness, and an isolated symbol extractor based on feature-vectors.<>