论文信息 - Analysis of 3-D scene with partially occluded objects for robot vision

Analysis of 3-D scene with partially occluded objects for robot vision

A vision system based on the attributed hypergraph representation (AHR) and monomorphism is presented. The graph representation consists of the structural descriptions of the objects in a scene and their interrelations. It renders all structural information explicit and symbolic. A hypergraph monomorphism algorithm is then used to compare the AHR of objects in the scene with a set of complete AHRs of prototypes. It enables objects which may be partially occluded by each other to be distinguished.<<ETX>>

Andrew K. C. Wong | Siwei Lu

[1] Andrew K. C. Wong,et al. Recognition of 3-D Objects in Range Images by Attributed Hypergraph Monomorphism and Synthesis , 1985 .

[2] Rangasami L. Kashyap,et al. Using Polygons to Recognize and Locate Partially Occluded Objects , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[3] Andrew K. C. Wong,et al. Synthesis of Attributed Hypergraphs for Knowledge Representation of 3-D Objects , 1988, Pattern Recognition.

[4] Robert C. Bolles,et al. Locating Partially Visible Objects: The Local Feature Focus Method , 1980, AAAI.

[5] Malcolm S. Casale,et al. An Overview of Analytic Solid Modeling , 1985, IEEE Computer Graphics and Applications.

[6] Norman I. Badler,et al. Decomposition of Three-Dimensional Objects into Spheres , 1979, IEEE Transactions on Pattern Analysis and Machine Intelligence.