1imageryintro(1) Grass User's Manual imageryintro(1)
2
6 Image data in general
7 In GRASS, image data are identical to raster data. However, a couple
8 of commands are explicitly dedicated to image processing. The geo‐
9 graphic boundaries of the raster/imagery file are described by the
10 north, south, east, and west fields. These values describe the lines
11 which bound the map at its edges. These lines do NOT pass through the
12 center of the grid cells at the edge of the map, but along the edge of
13 the map itself.
14 As a general rule in GRASS:
16 1
18 Raster/imagery output maps have their bounds and resolution
19 equal to those of the current region.
20 2
22 Raster/imagery input maps are automatically cropped/padded and
23 rescaled (using nearest-neighbor resampling) to match the cur‐
24 rent region.
25 Raster import
27 The module r.in.gdal offers a common interface for many different
28 raster and satellite image formats. Additionally, it also offers
29 options such as on-the-fly location creation or extension of the
30 default region to match the extent of the imported raster map. For
31 special cases, other import modules are available. Always the full map
32 is imported. Imagery data can be group (e.g. channel-wise) with
33 i.group.
34 For importing scanned maps, the user will need to create a x,y-loca‐
36 tion, scan the map in the desired resolution and save it into an appro‐
37 priate raster format (e.g. tiff, jpeg, png, pbm) and then use r.in.gdal
38 to import it. Based on reference points the scanned map can be recti‐
39 fied to obtain geocoded data.
40 Image processing operations
42 GRASS raster/imagery map processing is always performed in the current
43 region settings (see g.region), i.e. the current region extent and cur‐
44 rent raster resolution is used. If the resolution differs from that of
45 the input raster map(s), on-the-fly resampling is performed (nearest
46 neighbor resampling). If this is not desired, the input map(s) has/have
47 to be resampled beforehand with one of the dedicated modules.
48 Geocoding of imagery data
50 GRASS is able to geocode raster and image data of various types:
51 unreferenced scanned maps by defining four corner points
53 (i.target, i.rectify)
54 unreferenced satellite data from optical and Radar sen‐
56 sors by defining a certain number of ground control
57 points (i.target, i.rectify)
58 orthophoto based on DEM: i.ortho.photo
60 digital handheld camera geocoding: modified procedure for
62 i.ortho.photo
63 Visualizing (true) color composites
65 To quickly combine the first three channels to a near natural color
66 image, the GRASS command d.rgb can be used or the graphical GIS manager
67 (gis.m). It assigns each channel to a color which is then mixed while
68 displayed. With a bit more work of tuning the grey scales of the chan‐
69 nels, nearly perfect colors can be achieved. Channel histograms can be
70 shown with d.histogram.
71 Calculation of vegetation indices
73 An example for indices derived from multispectral data is the NDVI
74 (normalized difference vegetation index). To study the vegetation sta‐
75 tus with NDVI, the Red and the Near Infrared channels (NIR) are taken
76 as used as input for simple map algebra in the GRASS command r.mapcalc
77 (ndvi = 1.0 * (nir - red)/(nir + red)). With r.colors an optimized
78 "ndvi" color table can be assigned afterward. Also other vegetation
79 indices can be generated likewise.
80 Calibration of thermal channel
82 The encoded digital numbers of a thermal infrared channel can be trans‐
83 formed to degree Celsius (or other temperature units) which represent
84 the temperature of the observed land surface. This requires a few alge‐
85 braic steps with r.mapcalc which are outlined in the literature to
86 apply gain and bias values from the image metadata.
87 Image classification
89 Single and multispectral data can be classified to user defined land
90 use/land cover classes. In case of a single channel, segmentation will
91 be used. GRASS supports the following methods:
92 Radiometric classification:
94 Unsupervised classification (i.cluster, i.maxlik)
96 using the Maximum Likelihood classification method
97 Supervised classification (i.gensig or i.maxlik)
99 using the Maximum Likelihood classification method
100 Combined radiometric/geometric (segmentation based) supervised
101 classification (i.gensigset, i.smap)
102 Kappa statistic can be calculated to validate the results (r.kappa).
103 Image fusion
105 In case of using multispectral data, improvements of the resolution can
106 be gained by merging the panchromatic channel with color channels.
107 GRASS provides the HIS (i.rgb.his, i.his.rgb) and the Brovey transform
108 (i.fusion.brovey) methods.
109 Time series processing
111 GRASS also offers support for time series processing (<a
112 href="r.series.html">r.series). Statistics can be derived from a set of
113 coregistered input maps such as multitemporal satellite data. The com‐
114 mon univariate statistics and also linear regression can be calculated.
115 See also
117 Introduction to GRASS 2D raster map processing
118 Introduction to GRASS 3D raster map (voxel) processing
120 Introduction to GRASS vector map processing
122 full index
124GRASS 6.2.2 imageryintro(1)