Object Relationship Modeling based on Bayesian Network Integration for Improving Object Detection Performance of Service Robots

摘要: Recently tile study that exploits visual information for services of robot in indoor environments is active. Conventional image processing approaches are based on the pre-defined geometric models, so their performances likely to decrease when they applied uncertain and dynamic environments. For this, diverse researches manage uncertainty knowledge improving recognition performance have been doing. In this paper we propose a Bayesian network modeling method predicting existence target objects occluded by other ones object detection service robots. The proposed makes relationship, it allows predict through observed ones. define design small size networks (primitive netqork), allow integrate them following situations. experiments performed verifying constructed model, shows accuracy 5 places.