1v.kernel(1)                   Grass User's Manual                  v.kernel(1)
2
3
4

NAME

6       v.kernel  - Generates a raster density map from vector points map.
7       Density  is computed using a moving kernel. Optionally generates a vec‐
8       tor density map on a vector network.
9

KEYWORDS

11       vector, kernel density, point density, heatmap, hotspot
12

SYNOPSIS

14       v.kernel
15       v.kernel --help
16       v.kernel [-oqnm]  input=name   [net=name]    [output=name]    [net_out‐
17       put=name]    radius=float    [dsize=float]     [segmax=float]    [dist‐
18       max=float]     [multiplier=float]     [node=string]     [kernel=string]
19       [--overwrite]  [--help]  [--verbose]  [--quiet]  [--ui]
20
21   Flags:
22       -o
23           Try  to  calculate  an  optimal radius with given ’radius’ taken as
24           maximum (experimental)
25
26       -q
27           Only calculate optimal radius and exit (no map is written)
28
29       -n
30           In network mode, normalize values by sum of density  multiplied  by
31           length of each segment. Integral over the output map then gives 1.0
32           * mult
33
34       -m
35           In network mode, multiply the result by number of input points
36
37       --overwrite
38           Allow output files to overwrite existing files
39
40       --help
41           Print usage summary
42
43       --verbose
44           Verbose module output
45
46       --quiet
47           Quiet module output
48
49       --ui
50           Force launching GUI dialog
51
52   Parameters:
53       input=name [required]
54           Name of input vector map with training points
55
56       net=name
57           Name of input network vector map
58
59       output=name
60           Name for output raster map
61
62       net_output=name
63           Name for output vector density map
64           Outputs vector map if network map is given
65
66       radius=float [required]
67           Kernel radius in map units
68
69       dsize=float
70           Discretization error in map units
71           Default: 0.
72
73       segmax=float
74           Maximum length of segment on network
75           Default: 100.
76
77       distmax=float
78           Maximum distance from point to network
79           Default: 100.
80
81       multiplier=float
82           Multiply the density result by this number
83           Default: 1.
84
85       node=string
86           Node method
87           Options: none, split
88           Default: none
89           none: No method applied at nodes with more than 2 arcs
90           split: Equal split (Okabe 2009) applied at nodes
91
92       kernel=string
93           Kernel function
94           Options: uniform,  triangular,  epanechnikov,  quartic,  triweight,
95           gaussian, cosine
96           Default: gaussian
97

DESCRIPTION

99       v.kernel generates a raster density map from vector points data using a
100       moving kernel. Available kernel density functions are uniform, triangu‐
101       lar,  epanechnikov,  quartic,  triweight,  gaussian, cosine, default is
102       gaussian.
103
104       The module can also generate a vector density map on a vector  network.
105       Conventional  kernel functions produce biased estimates by overestimat‐
106       ing the densities around network nodes, whereas the equal split  method
107       of  Okabe  et al. (2009) produces unbiased density estimates. The equal
108       split method uses the kernel function selected with the  kernel  option
109       and can be enabled with node=split.
110

NOTES

112       The  multiplier  option  is  needed to overcome the limitation that the
113       resulting density in case of a vector map output is stored as  category
114       (integer).  The  density result stored as category may be multiplied by
115       this number.
116
117       For the gaussian kernel, standard deviation for the  gaussian  function
118       is set to 1/4 of the radius.
119
120       With the -o flag (experimental) the command tries to calculate an opti‐
121       mal radius. The value of radius is taken as maximum value.  The  radius
122       is  calculated  based  on  the gaussian function, using ALL points, not
123       just those in the current region.
124

EXAMPLES

126       Compute density of points (using vector map of schools from North  Car‐
127       olina sample dataset):
128       g.region region=wake_30m
129       v.kernel input=schools_wake output=schools_density radius=5000 multiplier=1000000
130       r.colors map=schools_density color=bcyr
131       School density
132

KNOWN ISSUES

134       The  module  only  considers  the presence of points, but not (yet) any
135       attribute values.
136

REFERENCES

138           ·   Okabe, A., Satoh, T., Sugihara, K.  (2009).  A  kernel  density
139               estimation  method for networks, its computational method and a
140               GIS-based tool.  International Journal of Geographical Informa‐
141               tion Science, Vol 23(1), pp. 7-32.
142               DOI: 10.1080/13658810802475491
143

SEE ALSO

145       v.surf.rst
146
147       Overview: Interpolation and Resampling in GRASS GIS
148

AUTHORS

150       Stefano Menegon, ITC-irst, Trento, Italy
151       Radim Blazek (additional kernel density functions and network part)
152

SOURCE CODE

154       Available at: v.kernel source code (history)
155
156       Main  index  | Vector index | Topics index | Keywords index | Graphical
157       index | Full index
158
159       © 2003-2019 GRASS Development Team, GRASS GIS 7.8.2 Reference Manual
160
161
162
163GRASS 7.8.2                                                        v.kernel(1)
Impressum