Title :
Polyhedral object recognition using view densities
Author_Institution :
Dept. of Electr. Eng. & Comput. Sci., Stevens Inst. of Technol., Hoboken, NJ, USA
Abstract :
A method of representing the features of a polyhedral object based on probability density functions defined on observed features is presented. The measured lengths of the edges of a polyhedral object depend on the observation angle. The measured lengths of a combination of adjacent edges in an image have characteristic relationships and associated probabilities of observation. These facts are used to develop a decision scheme for identifying polyhedral objects. Results of experiments are also presented
Keywords :
decision theory; pattern recognition; probability; decision scheme; pattern recognition; polyhedral object; probability density functions; view densities; Cameras; Communication systems; Data mining; Equations; Length measurement; Noise measurement; Object recognition; Probability density function; Robots; Stochastic processes;
Conference_Titel :
Systems, Man, and Cybernetics, 1991. 'Decision Aiding for Complex Systems, Conference Proceedings., 1991 IEEE International Conference on
Conference_Location :
Charlottesville, VA
Print_ISBN :
0-7803-0233-8
DOI :
10.1109/ICSMC.1991.169670
