Handling Uncertainty in 3D Object Recognition Using Bayesian Networks

摘要: In this paper we show how the uncertainty within a 3d recognition process can be modeled using Bayesian nets. Reliable object features in terms of rims are introduced to allow robust industrial free-form objects. Dependencies between observed and objects net. An algorithm build net from set CAD models is introduced. recognition, entering evidence into reduces possible hypotheses. Furthermore, expected change joint probability distribution allows an integration decision reasoning propagation. The selection optimal, next action incorporated nets reduce uncertainty.