1r.fillnulls(1) Grass User's Manual r.fillnulls(1)
2
6 r.fillnulls - Fills no-data areas in raster maps using spline interpo‐
7 lation.
8
10 raster, surface, elevation, interpolation, splines, no-data filling
11
13 r.fillnulls
14 r.fillnulls --help
15 r.fillnulls input=name output=name method=string [tension=float]
16 [smooth=float] [edge=integer] [npmin=integer] [segmax=integer]
17 [lambda=float] [memory=integer] [--overwrite] [--help] [--ver‐
18 bose] [--quiet] [--ui]
19 Flags:
21 --overwrite
22 Allow output files to overwrite existing files
23 --help
25 Print usage summary
26 --verbose
28 Verbose module output
29 --quiet
31 Quiet module output
32 --ui
34 Force launching GUI dialog
35 Parameters:
37 input=name [required]
38 Name of input raster map
39 output=name [required]
41 Name for output raster map
42 method=string [required]
44 Interpolation method to use
45 Options: bilinear, bicubic, rst
46 Default: rst
47 tension=float
49 Spline tension parameter
50 Default: 40.
51 smooth=float
53 Spline smoothing parameter
54 Default: 0.1
55 edge=integer
57 Width of hole edge used for interpolation (in cells)
58 Options: 2-100
59 Default: 3
60 npmin=integer
62 Minimum number of points for approximation in a segment (>segmax)
63 Options: 2-10000
64 Default: 600
65 segmax=integer
67 Maximum number of points in a segment
68 Options: 2-10000
69 Default: 300
70 lambda=float
72 Tykhonov regularization parameter (affects smoothing)
73 Used in bilinear and bicubic spline interpolation
74 Default: 0.01
75 memory=integer
77 Maximum memory to be used (in MB)
78 Cache size for raster rows
79 Default: 300
80
82 r.fillnulls fills NULL pixels (no data areas) in input raster map and
83 stores filled data to a new output raster map. The fill areas are
84 interpolated from the no data area boundaries buffer using v.surf.rst
85 regularized spline interpolation with tension (method=rst) or
86 r.resamp.bspline cubic or linear spline interpolation with Tykhonov
87 regularization.
88
90 Each area boundary buffer is set to three times the map resolution to
91 get nominally three points around the edge. This way the algorithm
92 interpolates into the hole with a trained slope and curvature at the
93 edges, in order to avoid that such a flat plane is generated in a hole.
94 The width of edge area can be adjusted by changing the edge parameter.
95 During the interpolation following warning may occur when using the RST
97 method:
98 Warning: strip exists with insufficient data
100 Warning: taking too long to find points for interpolation--please
101 change the region to area where your points are
102 This warning is generated if large data holes exist within the surface.
104 As the idea of r.fillnulls is to fill such holes, the user may ignore
105 the warning. The interpolation will be continued. However, the user may
106 pay attention to below notes.
107 If interpolation fails, temporary raster and vector maps are left in
109 place to allow unfilled map hole (NULL area) identification and manual
110 repair.
111 When using the default RST method, the algorithm is based on v.surf.rst
113 regularized splines with tension interpolation module which interpo‐
114 lates the raster cell values for NULL data areas from the boundary val‐
115 ues of the NULL data area. An eventual raster MASK is respected during
116 the NULL data area(s) filling. The interpolated values are patched into
117 the NULL data area(s) of the input map and saved into a new raster map.
118 Otherwise, either the linear or cubic spline interpolation with
119 Tykhonov regularization can be selected (based on r.resamp.bspline).
120
122 Depending on the shape of the NULL data area(s) problems may occur due
123 to an insufficient number of input cell values for the interpolation
124 process. Most problems will occur if a NULL data area reaches a large
125 amount of the map boundary. The user will have to carefully check the
126 result using r.mapcalc (generating a difference map to the input map
127 and applying the "differences" color table with r.colors) and/or to
128 query individual cell values.
129 RST method stores temporary maps on hard disk. It will require at least
131 as much free space as one extra input raster map takes.
132
134 In this example, the SRTM elevation map in the North Carolina sample
135 dataset location is filtered for outlier elevation values; missing pix‐
136 els are then re-interpolated to obtain a complete elevation map:
137 g.region raster=elev_srtm_30m -p
138 d.mon wx0
139 d.histogram elev_srtm_30m
140 # remove SRTM outliers, i.e. SRTM below 50m (esp. lakes), leading to no data areas
141 r.mapcalc "elev_srtm_30m_filt = if(elev_srtm_30m < 50.0, null(), elev_srtm_30m)"
142 d.histogram elev_srtm_30m_filt
143 d.rast elev_srtm_30m_filt
144 # using the default RST method to fill these holes in DEM
145 r.fillnulls input=elev_srtm_30m_filt output=elev_srtm_30m_rst tension=20
146 # using the bilinear method to fill these holes in DEM
147 r.fillnulls input=elev_srtm_30m_filt output=elev_srtm_30m_bilin method=bilinear
148 d.histogram elev_srtm_30m_rst
149 d.rast elev_srtm_30m_rst
150 d.erase
151 d.histogram elev_srtm_30m_bilin
152 d.rast elev_srtm_30m_bilin
153 r.mapcalc "diff_rst_bilin = elev_srtm_30m_rst - elev_srtm_30m_bilin"
154 r.colors diff_rst_bilin color=differences
155 r.univar -e diff_rst_bilin
156 d.erase
157 d.rast diff_rst_bilin
158 d.legend diff_rst_bilin
159
161 · Mitas, L., Mitasova, H., 1999, Spatial Interpolation. In:
162 P.Longley, M.F. Goodchild, D.J. Maguire, D.W.Rhind (Eds.), Geo‐
163 graphical Information Systems: Principles, Techniques, Manage‐
164 ment and Applications, Wiley, pp.481-492
165 · Mitasova H., Mitas L., Brown W.M., D.P. Gerdes, I. Kosi‐
167 novsky, Baker, T.1995, Modeling spatially and temporally dis‐
168 tributed phenomena: New methods and tools for GRASS GIS. Inter‐
169 national Journal of GIS, 9 (4), special issue on Integrating
170 GIS and Environmental modeling, 433-446.
171 · Mitasova H. and Mitas L. 1993: Interpolation by Regularized
173 Spline with Tension: I. Theory and Implementation, Mathemati‐
174 cal Geology 25, 641-655.
175 · Mitasova H. and Hofierka L. 1993: Interpolation by Regularized
177 Spline with Tension: II. Application to Terrain Modeling and
178 Surface Geometry Analysis, Mathematical Geology 25, 657-667.
179
181 r.fill.dir, r.mapcalc, r.resamp.bspline, v.surf.rst
182
184 Markus Neteler, University of Hannover and Fondazione Edmund Mach
185 Improvement by Hamish Bowman, NZ
186
188 Available at: r.fillnulls source code (history)
189 Main index | Raster index | Topics index | Keywords index | Graphical
191 index | Full index
192 © 2003-2019 GRASS Development Team, GRASS GIS 7.8.2 Reference Manual
194GRASS 7.8.2 r.fillnulls(1)