Title :
Comparison of scene segmentations: SMAP, ECHO, and maximum likelihood
Author :
McCauley, James Darrell ; Engel, Bernard A.
Author_Institution :
Dept. of Agric. Eng., Purdue Univ., West Lafayette, IN, USA
fDate :
11/1/1995 12:00:00 AM
Abstract :
Sequential maximum a posteriori (SMAP) and the extraction and classification of homogeneous objects (ECHO), two spectral/spatial scene segmentation algorithms, were compared with traditional maximum likelihood (ML) estimation in a supervised classification of multispectral data. SMAP generalized better than both ECHO and ML. Significant differences were found in all mean class classification accuracies: SMAP>ECHO>ML
Keywords :
geophysical signal processing; geophysical techniques; image classification; image segmentation; infrared imaging; maximum likelihood estimation; optical information processing; remote sensing; ECHO; IR imaging; SMAP; extraction and classification of homogeneous objects; geophysical measurement technique; image classification; image processing; image segmentation; land surface; maximum likelihood; multispectral method; optical imaging; scene segmentation; sequential maximum a posteriori; spatial scene segmentation algorithm; spectral segmentation algorithm; supervised classification; terrain mapping; visible; Bayesian methods; Data mining; Image segmentation; Layout; Markov random fields; Maximum a posteriori estimation; Maximum likelihood estimation; Pixel; Recursive estimation; Testing;
Journal_Title :
Geoscience and Remote Sensing, IEEE Transactions on
