Title :
Combining range and intensity data with a hidden Markov model
Author :
Huseby, R.B. ; Hogasen, G.T. ; Storvik, G. ; Aas, K.
Author_Institution :
Norsk Regnesentral, Oslo, Norway
fDate :
30 Aug-3 Sep 1992
Abstract :
The paper treats the analysis of an industrial inspection problem, namely the segmentation and discrimination of similar-looking bottles based on a multispectral image consisting of both range and intensity data. A contextual pixel classification is performed using a whole line as neighborhood. The framework of hidden Markov models together with a fast algorithm from control engineering makes this possible. The method is compared to J. Haslett´s method (1985) for contextual classification, and performs significantly better
Keywords :
Markov processes; automatic optical inspection; image segmentation; bottles; contextual pixel classification; discrimination; hidden Markov model; industrial inspection; intensity data; multispectral image; range data; segmentation; Automata; Bayesian methods; Control engineering; Hidden Markov models; Image analysis; Image generation; Image segmentation; Multispectral imaging; Pixel; Stochastic processes;
Conference_Titel :
Pattern Recognition, 1992. Vol.II. Conference B: Pattern Recognition Methodology and Systems, Proceedings., 11th IAPR International Conference on
Conference_Location :
The Hague
Print_ISBN :
0-8186-2915-0
DOI :
10.1109/ICPR.1992.201737
