1r.kappa(1) GRASS GIS User's Manual r.kappa(1)
2
3
4
6 r.kappa - Calculates error matrix and kappa parameter for accuracy as‐
7 sessment of classification result.
8
10 raster, statistics, classification
11
13 r.kappa
14 r.kappa --help
15 r.kappa [-whm] classification=name reference=name [output=name] [ti‐
16 tle=string] [--overwrite] [--help] [--verbose] [--quiet] [--ui]
17
18 Flags:
19 -w
20 Wide report
21 132 columns (default: 80)
22
23 -h
24 No header in the report
25
26 -m
27 Print Matrix only
28
29 --overwrite
30 Allow output files to overwrite existing files
31
32 --help
33 Print usage summary
34
35 --verbose
36 Verbose module output
37
38 --quiet
39 Quiet module output
40
41 --ui
42 Force launching GUI dialog
43
44 Parameters:
45 classification=name [required]
46 Name of raster map containing classification result
47
48 reference=name [required]
49 Name of raster map containing reference classes
50
51 output=name
52 Name for output file containing error matrix and kappa
53 If not given write to standard output
54
55 title=string
56 Title for error matrix and kappa
57 Default: ACCURACY ASSESSMENT
58
60 r.kappa tabulates the error matrix of classification result by crossing
61 classified map layer with respect to reference map layer. Both overall
62 kappa (accompanied by its variance) and conditional kappa values are
63 calculated. This analysis program respects the current geographic re‐
64 gion and mask settings.
65
66 r.kappa calculates the error matrix of the two map layers and prepares
67 the table from which the report is to be created. kappa values for
68 overall and each classes are computed along with their variances. Also
69 percent of comission and omission error, total correct classified re‐
70 sult by pixel counts, total area in pixel counts and percentage of
71 overall correctly classified pixels are tabulated.
72
73 The report will be write to an output file which is in plain text for‐
74 mat and named by user at prompt of running the program.
75
76 The body of the report is arranged in panels. The classified result
77 map layer categories is arranged along the vertical axis of the table,
78 while the reference map layer categories along the horizontal axis.
79 Each panel has a maximum of 5 categories (9 if wide format) across the
80 top. In addition, the last column of the last panel reflects a cross
81 total of each column for each row. All of the categories of the map
82 layer arranged along the vertical axis, i.e., the reference map layer,
83 are included in each panel. There is a total at the bottom of each
84 column representing the sum of all the rows in that column.
85
87 It is recommended to reclassify categories of classified result map
88 layer into a more manageable number before running r.kappa on the clas‐
89 sified raster map layer. Because r.kappa calculates and then reports
90 information for each and every category.
91
92 NA’s in output file mean non-applicable in case MASK exists.
93
94 The Estimated kappa value in r.kappa is the value only for one class,
95 i.e. the observed agreement between the classifications for those ob‐
96 servations that have been classified by classifier 1 into the class i.
97 In other words, here the choice of reference is important.
98
99 It is calculated as:
100
101 kpp[i] = (pii[i] - pi[i] * pj[i]) / (pi[i] - pi[i] * pj[i]);
102
103 where=
104
105 • pii[i] is the probability of agreement (i.e. number of pixels
106 for which there is agreement divided by total number of as‐
107 sessed pixels)
108
109 • Pi[i] is the probability of classification i having classified
110 the point as i
111
112 • Pj[i] is the probability of classification j having classified
113 the point as i.
114
116 Example for North Carolina sample dataset:
117 g.region raster=landclass96 -p
118 r.kappa -w classification=landuse96_28m reference=landclass96
119 # export Kappa matrix as CSV file "kappa.csv"
120 r.kappa classification=landuse96_28m reference=landclass96 output=kappa.csv -m -h
121
122 Verification of classified LANDSAT scene against training areas:
123 r.kappa -w classification=lsat7_2002_classes reference=training
124
126 g.region, r.category, r.mask, r.reclass, r.report, r.stats
127
129 Tao Wen, University of Illinois at Urbana-Champaign, Illinois
130
132 Available at: r.kappa source code (history)
133
134 Accessed: Saturday Jan 21 21:14:45 2023
135
136 Main index | Raster index | Topics index | Keywords index | Graphical
137 index | Full index
138
139 © 2003-2023 GRASS Development Team, GRASS GIS 8.2.1 Reference Manual
140
141
142
143GRASS 8.2.1 r.kappa(1)