Junction Detection with Automatic Selection of Detection Scales and Localization Scales

The subject of scale selection is essential to many aspects of multi-scale and multi-resolution processing of image data. This article shows how a general heuristic principle for scale selection can be applied to the problem of detecting and localizing junctions. In a rst uncommitted processing step initial hypotheses about interesting scale levels (and regions of interest) are generated from scales where normalized di erential invariants assume maxima over scales (and space). Then, based on this scale (and region) information, a more re ned processing stage is invoked tuned to the task at hand. The resulting method is the rst junction detector with automatic scale selection. Whereas this article deals with the speci c problem of junction detection, the underlying ideas apply also to other types of di erential feature detectors, such as blob detectors, edge detectors, and ridge detectors.