A Symmetric Convexity Measure

A new area-based convexity measure for polygons is described. It has the desirable properties that it is not sensitive to small boundary defects, and it is symmetric with respect to intrusions and protrusions. The measure requires a maximally overlapping convex polygon, and this is efficiently estimated using a genetic algorithm. Examples of the measures application to medical image analysis are shown.

• Type: Conference
• Year: 2004
• Where: ICPR
• Authors: Paul L. Rosin, Christine L. Mumford
• Keywords: Area-based Convexity Measure, Computer Vision, ICPR 2004, Overlapping Convex Polygon, Small Boundary Defects