Abstract
We propose a method for recognizing 3-D objects by use of images taken at several fixed viewpoints. In the method, we suppose that each input image possesses an agent which has object models in its model database and can execute feature extraction, reconstruction and model matching processes. Under such a supposition, each agent is intended to implement the analysis on the associated image. In this processing level, however, the method can’t recognize objects, and even if it can derive some results, they aie possibly unreliable because of the restriction of view directions or occlusions. Therefore, it is necessary to integrate information from other images to derive a correct solution.
We introduce a new mechanism in which the agent being able to provide the largest amount of valuable information about the observed objects will be chosen as the supervisor and play the role of integrating results from other agents. However, it is difficult to formulate rules for evaluating if the chosen agent can produce a good result for the recognition, because various factors should be combined for the evaluation. To resolve the difficulty, we use the analytic hierarchy process (AHP) as the evaluation rules.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
T.Kojoh, T.Nagata, H.B.Zha, Cooperative Recognition using Multi-View Images Proceedings of Korea Automatic Control Conference (′93 KACC, Int. Session), 1993, pp.70–75
M.Ito, A.Iskii, Three View Stereo Analysis, IEEE Trans. PAMI, Vol.8, No4, 1986, pp.524–533
R.Chiou, K.Hung, J.Guo, C.Chen, T.Fan, J.Lee, Polyhedron Recognition Using Three-view Analysis, Pattern Recognition, Vol.25, No.1, 1992, pp. 1–16
R.Mohan, D.Weinshall, R.R.Sarukkai, 3D Object Recognition By Indexing Structural Invariants from Multiple Views, Proceedings of IEEE ICCV, 1993, pp.264–269
Z.Chen, S.Ho, Incremental Model Building of Polyhedral Invariants from Multiple View, Pattern Recognition, Vol.26 No.1, 1993, pp.33–46
Y.C Tang, C.S George Lee, A Geometric Feature Relation Graph Formulation for Consistent Sensor Fusion, IEEE Trans. PAMI, Vol.22, No.1, 1992, pp.115–129
B.Besser, S.Estable, B.Ulmer, Multiple Knowledge Sources and Evidential Reasoning for Shape Recognition, Proceedings of IEEE ICCV, 1993, pp.624–631
B.A.Draper, J.B.Brolio, T.Collins, A.R.Hanson, E.M Riseman, Image Interpretation by Distributed Cooperative Process, Proceedings of IEEE CVPR, 1988, pp.129–135
E.W.Kent, M.O.Shneier, T.H.Hong, Building Representations from Fusions of Multiple Views, Proceedings of IEEE CVPR, 1998, pp.1634–1639
T.L.Saaty, The Analytic Hierarchy Process, McGraw-Hill, 1981
T.L.Saaty, L.G. Vargas, Prediction Projection and Forecasting, Kiuwer Academic Publishers, 1991
P.J.Woodham, Analyzing Images of Curved Surface, Artificial Intelligence, Vol.17, 1981, pp.117–140
R.O. Duda, Use of the Hough Transformation to Detect Line and Curves in Pictures, Comm.ACM, Vol.15, No.1, 1972, pp.11–15
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 1994 Springer-Verlag Tokyo
About this paper
Cite this paper
Kojoh, T., Nagata, T., Zha, HB. (1994). Fusing Image Information on the Basis of the Analytic Hierarchy Process. In: Asama, H., Fukuda, T., Arai, T., Endo, I. (eds) Distributed Autonomous Robotic Systems. Springer, Tokyo. https://doi.org/10.1007/978-4-431-68275-2_8
Download citation
DOI: https://doi.org/10.1007/978-4-431-68275-2_8
Publisher Name: Springer, Tokyo
Print ISBN: 978-4-431-68277-6
Online ISBN: 978-4-431-68275-2
eBook Packages: Springer Book Archive