Title :
Unsupervised multispectral image classification using MRF models and VQ method
Author :
Yamazaki, Tatsuya ; Gingras, Denis
Author_Institution :
Kansai Adv. Res. Center, Minist. of Posts & Telecommun., Tokyo, Japan
fDate :
3/1/1999 12:00:00 AM
Abstract :
An unsupervised contextual classification method using Markov random field (MRF) models and the vector quantization (VQ) is proposed. The VQ algorithm classifies the observed data preliminarily, and the contextual reclassification and the iterative classification follow. The proposed method was applied to a Landsat image to evaluate the classification accuracy
Keywords :
Markov processes; geophysical signal processing; geophysical techniques; image classification; iterative methods; multidimensional signal processing; random processes; remote sensing; terrain mapping; vector quantisation; Landsat image; MRF model; Markov random field; VQ algorithm; VQ method; context; contextual classification method; geophysical measurement technique; image classification; iterative classification; land surface; multispectral remote sensing; optical imaging; reclassification; terrain mapping; unsupervised classification; vector quantization; Context modeling; Frequency; Image classification; Isolation technology; Iterative algorithms; Markov random fields; Multispectral imaging; Remote sensing; Satellites; Vector quantization;
Journal_Title :
Geoscience and Remote Sensing, IEEE Transactions on
