Fuzzy belief and scene interpretation
Sandy Dance, Zhi-Qiang Liu · 2002
We present an image interpretation system using fuzzy belief measures. This system is based on a symbolic object-oriented image interpretation system (SOO-PIN) we developed previously. In many image analysis and understanding applications, objects can only be described vaguely. Using fuzzy belief measures and linguistic hedges we are able to manage uncertainty properly. Furthermore, this enables us to develop an image understanding system that produces a natural linguistic presentation and provides a user-friendly environment. We demonstrate the effectiveness of our system with a complex traffic scene.