作者: Enrico Costanza , John Robinson
DOI:
关键词: Data mining 、 Adjacency list 、 Algorithm 、 Tree (data structure) 、 Computer science 、 Independence (probability theory) 、 Fiducial marker 、 Distortion 、 Set (abstract data type)
摘要: We report a topological approach to fiducial recognition for real-time applications. Independence from geometry makes the system tolerant severe distortion, and allows encoding of extra information. The method is based on region adjacency trees. After describing mathematical foundations, we present set simulations evaluate algorithm optimise design.