i.smap.1grass - Man Page

imagery, classification, supervised classification, segmentation, SMAP

i.smap
i.smap --help
i.smap [-m] group=name subgroup=name signaturefile=name output=name  [goodness=name]   [blocksize=integer]   [--overwrite]  [--help]  [--verbose]  [--quiet]  [--ui]

The i.smap program is used to segment multispectral images using a spectral class model known as a Gaussian mixture distribution.  Since Gaussian mixture distributions include conventional multivariate Gaussian distributions, this program may also be used to segment multispectral images based on simple spectral mean and covariance parameters.

i.smap has two modes of operation. The first mode is the sequential maximum a posteriori (SMAP) mode [1,2].  The SMAP segmentation algorithm attempts to improve segmentation accuracy by segmenting the image into regions rather than segmenting each pixel separately (see Notes).

The second mode is the more conventional maximum likelihood (ML) classification which classifies each pixel separately, but requires somewhat less computation. This mode is selected with the -m flag (see below).

The SMAP algorithm exploits the fact that nearby pixels in an image are likely to have the same class.  It works by segmenting the image at various scales or resolutions and using the coarse scale segmentations to guide the finer scale segmentations.  In addition to reducing the number of misclassifications, the SMAP algorithm generally produces segmentations with larger connected regions of a fixed class which may be useful in some applications.

The amount of smoothing that is performed in the segmentation is dependent of the behaviour of the data in the image.  If the data suggests that the nearby pixels often change class, then the algorithm will adaptively reduce the amount of smoothing.  This ensures that excessively large regions are not formed.

The degree of misclassifications can be investigated with the goodness of fit output map. Lower values indicate a better fit. The largest 5 to 15% of the goodness values may need some closer inspection.

The module i.smap does not support MASKed or NULL cells. Therefore it might be necessary to create a copy of the classification results using e.g. r.mapcalc:

r.mapcalc "MASKed_map = classification_results"

Supervised classification of LANDSAT scene (complete NC dataset)

# Align computation region to the scene
g.region raster=lsat7_2002_10 -p
# store VIZ, NIR, MIR into group/subgroup
i.group group=lsat7_2002 subgroup=res_30m \
  input=lsat7_2002_10,lsat7_2002_20,lsat7_2002_30,lsat7_2002_40,lsat7_2002_50,lsat7_2002_70
# Now digitize training areas "training" with the digitizer
# and convert to raster model with v.to.rast
v.to.rast input=training output=training use=cat label_column=label
# If you are just playing around and do not care about the accuracy of outcome,
# just use one of existing maps instead e.g.
# g.copy rast=landuse96_28m,training
# Create a signature file with statistics for each class
i.gensigset trainingmap=training group=lsat7_2002 subgroup=res_30m \
            signaturefile=lsat7_2002_30m maxsig=5
# Predict classes based on whole LANDSAT scene
i.smap group=lsat7_2002 subgroup=res_30m signaturefile=lsat7_2002_30m \
       output=lsat7_2002_smap_classes
# Visually check result
d.mon wx0
d.rast.leg lsat7_2002_smap_classes
# Statistically check result
r.kappa -w classification=lsat7_2002_smap_classes reference=training

The signature file obtained in the example above will allow to classify the current imagery group only (lsat7_2002). If the user would like to re-use the signature file for the classification of different imagery group(s), they can set semantic labels for each group member beforehand, i.e., before generating the signature files. Semantic labels are set by means of r.support as shown below:

# Define semantic labels for all LANDSAT bands
r.support map=lsat7_2002_10 semantic_label=TM7_1
r.support map=lsat7_2002_20 semantic_label=TM7_2
r.support map=lsat7_2002_30 semantic_label=TM7_3
r.support map=lsat7_2002_40 semantic_label=TM7_4
r.support map=lsat7_2002_50 semantic_label=TM7_5
r.support map=lsat7_2002_61 semantic_label=TM7_61
r.support map=lsat7_2002_62 semantic_label=TM7_62
r.support map=lsat7_2002_70 semantic_label=TM7_7
r.support map=lsat7_2002_80 semantic_label=TM7_8

• C. Bouman and M. Shapiro, "Multispectral Image Segmentation using a Multiscale Image Model", Proc. of IEEE Int’l Conf. on Acoust., Speech and Sig. Proc., pp. III-565 - III-568, San Francisco, California, March 23-26, 1992.
• C. Bouman and M. Shapiro 1994, "A Multiscale Random Field Model for Bayesian Image Segmentation", IEEE Trans. on Image Processing., 3(2), 162-177" (PDF)
• McCauley, J.D. and B.A. Engel 1995, "Comparison of Scene Segmentations: SMAP, ECHO and Maximum Likelihood", IEEE Trans. on Geoscience and Remote Sensing, 33(6): 1313-1316.

r.support for setting semantic labels,
i.group for creating groups and subgroups
r.mapcalc to copy classification result in order to cut out MASKed subareas
i.gensigset to generate the signature file required by this program

g.gui.iclass, i.maxlik, r.kappa

Charles Bouman, School of Electrical Engineering, Purdue University
Michael Shapiro, U.S.Army Construction Engineering Research Laboratory
Semantic label support: Maris Nartiss, University of Latvia

Available at: i.smap source code (history)

Accessed: Monday Sep 09 08:49:43 2024

Main index | Imagery index | Topics index | Keywords index | Graphical index | Full index

© 2003-2024 GRASS Development Team, GRASS GIS 8.4.0 Reference Manual