Detailled semantic labeling (2D) result


Name H. Wang
Affiliation NLPR, Inst. of Automation, China
Abbreviation GSN3
Strategy (u)nsupervised, (s)upervised, (h)ybrid s


Overall statistics, reference set: full_reference


Number of non processed tiles by this participant: 0
↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.912
0.025
0.042
0.017
0.005
0.000
building
0.049
0.925
0.020
0.005
0.001
0.000
low_veg
0.056
0.023
0.778
0.142
0.000
0.000
tree
0.020
0.004
0.083
0.893
0.000
0.000
car
0.258
0.034
0.006
0.005
0.698
0.000
clutter
0.395
0.234
0.021
0.007
0.050
0.293
Precision/Correctness
0.875
0.940
0.825
0.846
0.813
0.983
Recall/Completeness
0.912
0.925
0.778
0.893
0.698
0.293
F1
0.893
0.933
0.801
0.869
0.751
0.451

Overall accuracy 0.874

Overall statistics, reference set: no_boundary


Number of non processed tiles by this participant: 0
↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.937
0.019
0.034
0.008
0.003
0.000
building
0.040
0.944
0.013
0.002
0.001
0.000
low_veg
0.044
0.016
0.814
0.126
0.000
0.000
tree
0.009
0.002
0.066
0.923
0.000
0.000
car
0.152
0.038
0.003
0.002
0.804
0.000
clutter
0.356
0.249
0.019
0.002
0.052
0.322
Precision/Correctness
0.907
0.957
0.860
0.876
0.845
0.993
Recall/Completeness
0.937
0.944
0.814
0.923
0.804
0.322
F1
0.922
0.951
0.837
0.899
0.824
0.487

Overall accuracy 0.903
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9464
0.0119
0.0238
0.0126
0.0054
0.0000
building
0.0384
0.9367
0.0220
0.0028
0.0000
0.0000
low_veg
0.0674
0.0149
0.7327
0.1848
0.0003
0.0000
tree
0.0245
0.0038
0.0793
0.8909
0.0001
0.0014
car
0.2527
0.0060
0.0003
0.0013
0.7398
0.0000
clutter
0.0005
0.0000
0.0000
0.0091
0.0000
0.9904
Precision/Correctness
0.870
0.966
0.854
0.817
0.889
0.981
Recall/Completeness
0.946
0.937
0.733
0.891
0.740
0.990
F1
0.907
0.951
0.789
0.853
0.807
0.986

Overall accuracy 0.876

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9709
0.0086
0.0134
0.0034
0.0037
0.0000
building
0.0308
0.9563
0.0126
0.0003
0.0000
0.0000
low_veg
0.0508
0.0081
0.7625
0.1783
0.0003
0.0000
tree
0.0104
0.0009
0.0655
0.9226
0.0000
0.0006
car
0.1215
0.0041
0.0000
0.0004
0.8740
0.0000
clutter
0.0002
0.0000
0.0000
0.0013
0.0000
0.9985
Precision/Correctness
0.911
0.981
0.890
0.840
0.915
0.993
Recall/Completeness
0.971
0.956
0.763
0.923
0.874
0.998
F1
0.940
0.969
0.821
0.880
0.894
0.996

Overall accuracy 0.905

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9486
0.0297
0.0130
0.0051
0.0035
0.0000
building
0.0729
0.9159
0.0088
0.0017
0.0007
0.0000
low_veg
0.1253
0.0603
0.5043
0.3093
0.0008
0.0000
tree
0.0417
0.0116
0.0482
0.8980
0.0005
0.0000
car
0.3378
0.0107
0.0009
0.0037
0.6469
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.838
0.977
0.670
0.725
0.864
--
Recall/Completeness
0.949
0.916
0.504
0.898
0.647
--
F1
0.890
0.945
0.575
0.803
0.740
--

Overall accuracy 0.902

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9731
0.0138
0.0097
0.0014
0.0020
0.0000
building
0.0591
0.9320
0.0078
0.0004
0.0006
0.0000
low_veg
0.0886
0.0360
0.5402
0.3348
0.0005
0.0000
tree
0.0165
0.0057
0.0417
0.9360
0.0001
0.0000
car
0.2118
0.0070
0.0000
0.0021
0.7792
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.870
0.990
0.699
0.757
0.899
--
Recall/Completeness
0.973
0.932
0.540
0.936
0.779
--
F1
0.919
0.960
0.609
0.837
0.835
--

Overall accuracy 0.926

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area6_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9218
0.0375
0.0256
0.0109
0.0042
0.0000
building
0.0510
0.9311
0.0138
0.0037
0.0004
0.0000
low_veg
0.0790
0.0196
0.7177
0.1833
0.0004
0.0000
tree
0.0156
0.0022
0.0505
0.9316
0.0001
0.0000
car
0.2919
0.0092
0.0007
0.0015
0.6967
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.872
0.961
0.749
0.919
0.898
--
Recall/Completeness
0.922
0.931
0.718
0.932
0.697
--
F1
0.896
0.946
0.733
0.925
0.785
--

Overall accuracy 0.902

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9540
0.0251
0.0161
0.0031
0.0017
0.0000
building
0.0399
0.9498
0.0090
0.0009
0.0003
0.0000
low_veg
0.0649
0.0078
0.7631
0.1642
0.0000
0.0000
tree
0.0074
0.0004
0.0388
0.9535
0.0000
0.0000
car
0.1748
0.0047
0.0001
0.0000
0.8205
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.907
0.978
0.799
0.943
0.950
--
Recall/Completeness
0.954
0.950
0.763
0.953
0.820
--
F1
0.930
0.964
0.781
0.948
0.880
--

Overall accuracy 0.932

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8884
0.0266
0.0708
0.0103
0.0039
0.0000
building
0.0446
0.9334
0.0173
0.0040
0.0007
0.0000
low_veg
0.0379
0.0114
0.8655
0.0843
0.0009
0.0000
tree
0.0182
0.0052
0.0726
0.9039
0.0001
0.0000
car
0.2552
0.0102
0.0139
0.0040
0.7166
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.913
0.947
0.795
0.912
0.885
--
Recall/Completeness
0.888
0.933
0.865
0.904
0.717
--
F1
0.901
0.940
0.829
0.908
0.792
--

Overall accuracy 0.895

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9137
0.0214
0.0601
0.0028
0.0020
0.0000
building
0.0336
0.9554
0.0091
0.0013
0.0006
0.0000
low_veg
0.0216
0.0049
0.8992
0.0740
0.0003
0.0000
tree
0.0061
0.0024
0.0523
0.9392
0.0000
0.0000
car
0.1423
0.0100
0.0090
0.0027
0.8361
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.948
0.965
0.838
0.935
0.929
--
Recall/Completeness
0.914
0.955
0.899
0.939
0.836
--
F1
0.931
0.960
0.868
0.937
0.880
--

Overall accuracy 0.926

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8620
0.0477
0.0678
0.0187
0.0038
0.0000
building
0.0401
0.9350
0.0142
0.0102
0.0004
0.0001
low_veg
0.0819
0.0226
0.7432
0.1523
0.0000
0.0000
tree
0.0113
0.0067
0.0765
0.9054
0.0000
0.0000
car
0.2144
0.0361
0.0008
0.0031
0.7456
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.921
0.904
0.746
0.837
0.776
--
Recall/Completeness
0.862
0.935
0.743
0.905
0.746
--
F1
0.891
0.919
0.745
0.870
0.760
--

Overall accuracy 0.866

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8936
0.0367
0.0597
0.0073
0.0028
0.0000
building
0.0333
0.9490
0.0103
0.0070
0.0003
0.0001
low_veg
0.0793
0.0107
0.7861
0.1239
0.0000
0.0000
tree
0.0045
0.0031
0.0561
0.9364
0.0000
0.0000
car
0.1138
0.0414
0.0000
0.0001
0.8448
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.937
0.932
0.784
0.880
0.797
--
Recall/Completeness
0.894
0.949
0.786
0.936
0.845
--
F1
0.915
0.940
0.785
0.908
0.820
--

Overall accuracy 0.897

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8959
0.0259
0.0470
0.0281
0.0030
0.0002
building
0.0294
0.9534
0.0128
0.0044
0.0000
0.0000
low_veg
0.0502
0.0191
0.7705
0.1599
0.0003
0.0000
tree
0.0118
0.0059
0.0952
0.8871
0.0000
0.0000
car
0.1969
0.0000
0.0000
0.0069
0.7962
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.833
0.941
0.834
0.865
0.888
--
Recall/Completeness
0.896
0.953
0.770
0.887
0.796
--
F1
0.863
0.947
0.801
0.876
0.840
--

Overall accuracy 0.867

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9222
0.0239
0.0386
0.0137
0.0014
0.0002
building
0.0267
0.9670
0.0053
0.0010
0.0000
0.0000
low_veg
0.0473
0.0118
0.8060
0.1348
0.0002
0.0000
tree
0.0067
0.0017
0.0793
0.9123
0.0000
0.0000
car
0.0679
0.0000
0.0000
0.0014
0.9306
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.861
0.962
0.864
0.894
0.928
--
Recall/Completeness
0.922
0.967
0.806
0.912
0.931
--
F1
0.890
0.965
0.834
0.903
0.929
--

Overall accuracy 0.894

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8305
0.0275
0.1024
0.0337
0.0059
0.0000
building
0.0221
0.9609
0.0120
0.0050
0.0000
0.0000
low_veg
0.0835
0.0456
0.7057
0.1643
0.0009
0.0000
tree
0.0267
0.0099
0.1098
0.8536
0.0001
0.0000
car
0.1581
0.0000
0.0001
0.0033
0.8386
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.847
0.888
0.742
0.850
0.583
--
Recall/Completeness
0.831
0.961
0.706
0.854
0.839
--
F1
0.839
0.923
0.723
0.852
0.688
--

Overall accuracy 0.829

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area14_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8513
0.0227
0.0979
0.0230
0.0052
0.0000
building
0.0165
0.9774
0.0052
0.0009
0.0000
0.0000
low_veg
0.0783
0.0309
0.7365
0.1535
0.0009
0.0000
tree
0.0192
0.0039
0.0965
0.8803
0.0000
0.0000
car
0.0486
0.0000
0.0000
0.0000
0.9514
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.873
0.923
0.765
0.872
0.568
--
Recall/Completeness
0.851
0.977
0.737
0.880
0.951
--
F1
0.862
0.949
0.750
0.876
0.711
--

Overall accuracy 0.855

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8185
0.0276
0.1013
0.0502
0.0024
0.0000
building
0.0460
0.9184
0.0263
0.0088
0.0005
0.0000
low_veg
0.0253
0.0112
0.8542
0.1091
0.0000
0.0001
tree
0.0219
0.0037
0.1118
0.8625
0.0000
0.0000
car
0.1447
0.0011
0.0171
0.0058
0.8313
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.840
0.948
0.805
0.884
0.794
--
Recall/Completeness
0.819
0.918
0.854
0.863
0.831
--
F1
0.829
0.933
0.829
0.873
0.812
--

Overall accuracy 0.863

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8527
0.0239
0.0924
0.0303
0.0008
0.0000
building
0.0413
0.9419
0.0136
0.0028
0.0004
0.0000
low_veg
0.0213
0.0064
0.8784
0.0937
0.0000
0.0001
tree
0.0147
0.0013
0.0963
0.8878
0.0000
0.0000
car
0.0805
0.0000
0.0050
0.0004
0.9142
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.870
0.966
0.834
0.908
0.885
--
Recall/Completeness
0.853
0.942
0.878
0.888
0.914
--
F1
0.861
0.954
0.856
0.898
0.899
--

Overall accuracy 0.889

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8586
0.0377
0.0481
0.0517
0.0040
0.0000
building
0.0177
0.9648
0.0117
0.0058
0.0000
0.0000
low_veg
0.0601
0.0253
0.7330
0.1807
0.0006
0.0004
tree
0.0072
0.0011
0.0421
0.9495
0.0000
0.0000
car
0.0791
0.0337
0.0102
0.0132
0.8638
0.0000
clutter
0.1016
0.8953
0.0015
0.0016
0.0000
0.0000
Precision/Correctness
0.894
0.809
0.830
0.894
0.794
0.000
Recall/Completeness
0.859
0.965
0.733
0.950
0.864
0.000
F1
0.876
0.880
0.779
0.921
0.827
-nan

Overall accuracy 0.866

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8904
0.0313
0.0442
0.0322
0.0019
0.0000
building
0.0142
0.9764
0.0081
0.0013
0.0000
0.0000
low_veg
0.0532
0.0155
0.7777
0.1527
0.0006
0.0004
tree
0.0030
0.0001
0.0301
0.9668
0.0000
0.0000
car
0.0116
0.0391
0.0000
0.0067
0.9426
0.0000
clutter
0.0908
0.9087
0.0000
0.0005
0.0000
0.0000
Precision/Correctness
0.918
0.823
0.863
0.920
0.850
0.000
Recall/Completeness
0.890
0.976
0.778
0.967
0.943
0.000
F1
0.904
0.893
0.818
0.943
0.894
-nan

Overall accuracy 0.891

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9007
0.0200
0.0393
0.0333
0.0066
0.0000
building
0.0299
0.9540
0.0090
0.0067
0.0003
0.0000
low_veg
0.0900
0.0271
0.7301
0.1518
0.0010
0.0000
tree
0.0193
0.0035
0.0469
0.9303
0.0000
0.0000
car
0.2262
0.0170
0.0083
0.0126
0.7359
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.901
0.929
0.841
0.863
0.773
--
Recall/Completeness
0.901
0.954
0.730
0.930
0.736
--
F1
0.901
0.941
0.782
0.895
0.754
--

Overall accuracy 0.882

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9325
0.0139
0.0328
0.0164
0.0045
0.0000
building
0.0234
0.9711
0.0025
0.0028
0.0002
0.0000
low_veg
0.0861
0.0146
0.7767
0.1217
0.0010
0.0000
tree
0.0077
0.0009
0.0319
0.9595
0.0000
0.0000
car
0.1184
0.0209
0.0040
0.0045
0.8522
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.926
0.958
0.882
0.903
0.805
--
Recall/Completeness
0.932
0.971
0.777
0.960
0.852
--
F1
0.929
0.964
0.826
0.930
0.828
--

Overall accuracy 0.916

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8284
0.0252
0.1064
0.0335
0.0065
0.0000
building
0.0764
0.8557
0.0506
0.0173
0.0000
0.0000
low_veg
0.0202
0.0123
0.8867
0.0804
0.0003
0.0000
tree
0.0217
0.0055
0.1426
0.8302
0.0000
0.0000
car
0.1712
0.0027
0.0006
0.0019
0.8236
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.807
0.941
0.856
0.836
0.741
--
Recall/Completeness
0.828
0.856
0.887
0.830
0.824
--
F1
0.818
0.896
0.871
0.833
0.780
--

Overall accuracy 0.858

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8598
0.0155
0.0972
0.0227
0.0047
0.0000
building
0.0717
0.8690
0.0450
0.0142
0.0000
0.0000
low_veg
0.0166
0.0083
0.9057
0.0691
0.0003
0.0000
tree
0.0151
0.0025
0.1208
0.8616
0.0000
0.0000
car
0.0623
0.0033
0.0000
0.0000
0.9344
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.834
0.963
0.878
0.861
0.758
--
Recall/Completeness
0.860
0.869
0.906
0.862
0.934
--
F1
0.846
0.914
0.892
0.861
0.837
--

Overall accuracy 0.881

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9388
0.0259
0.0199
0.0054
0.0100
0.0000
building
0.0715
0.9093
0.0154
0.0035
0.0004
0.0000
low_veg
0.0837
0.0274
0.6967
0.1915
0.0007
0.0000
tree
0.0123
0.0045
0.0518
0.9314
0.0001
0.0000
car
0.2956
0.0043
0.0069
0.0043
0.6888
0.0000
clutter
0.4340
0.3052
0.1404
0.0087
0.1117
0.0000
Precision/Correctness
0.849
0.955
0.886
0.697
0.857
-nan
Recall/Completeness
0.939
0.909
0.697
0.931
0.689
0.000
F1
0.892
0.932
0.780
0.797
0.764
-nan

Overall accuracy 0.870

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9647
0.0167
0.0131
0.0017
0.0038
0.0000
building
0.0590
0.9291
0.0107
0.0009
0.0003
0.0000
low_veg
0.0648
0.0142
0.7415
0.1790
0.0005
0.0000
tree
0.0042
0.0011
0.0366
0.9581
0.0000
0.0000
car
0.1994
0.0030
0.0052
0.0035
0.7889
0.0000
clutter
0.4441
0.2908
0.1215
0.0061
0.1376
0.0000
Precision/Correctness
0.885
0.975
0.921
0.729
0.921
-nan
Recall/Completeness
0.965
0.929
0.741
0.958
0.789
0.000
F1
0.923
0.951
0.822
0.828
0.850
-nan

Overall accuracy 0.902

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9452
0.0162
0.0199
0.0132
0.0054
0.0000
building
0.0321
0.9488
0.0119
0.0071
0.0000
0.0000
low_veg
0.0534
0.0175
0.6261
0.3025
0.0005
0.0000
tree
0.0036
0.0020
0.0204
0.9740
0.0000
0.0000
car
0.2613
0.0031
0.0067
0.0099
0.7190
0.0000
clutter
0.5630
0.3433
0.0210
0.0077
0.0599
0.0052
Precision/Correctness
0.915
0.928
0.864
0.669
0.699
0.958
Recall/Completeness
0.945
0.949
0.626
0.974
0.719
0.005
F1
0.930
0.938
0.726
0.793
0.709
0.010

Overall accuracy 0.871

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9663
0.0107
0.0136
0.0054
0.0040
0.0000
building
0.0254
0.9642
0.0070
0.0034
0.0000
0.0000
low_veg
0.0328
0.0072
0.6695
0.2902
0.0002
0.0000
tree
0.0007
0.0003
0.0137
0.9853
0.0000
0.0000
car
0.1617
0.0024
0.0068
0.0072
0.8219
0.0000
clutter
0.5490
0.3622
0.0140
0.0029
0.0657
0.0062
Precision/Correctness
0.937
0.946
0.904
0.708
0.703
1.000
Recall/Completeness
0.966
0.964
0.670
0.985
0.822
0.006
F1
0.951
0.955
0.769
0.824
0.758
0.012

Overall accuracy 0.899

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9770
0.0050
0.0090
0.0039
0.0050
0.0002
building
0.0501
0.9338
0.0095
0.0025
0.0041
0.0000
low_veg
0.0451
0.0028
0.9144
0.0370
0.0006
0.0001
tree
0.0322
0.0013
0.0702
0.8955
0.0007
0.0000
car
0.1797
0.1676
0.0041
0.0036
0.6451
0.0000
clutter
0.5594
0.0994
0.0646
0.0156
0.0812
0.1798
Precision/Correctness
0.885
0.956
0.893
0.924
0.796
0.986
Recall/Completeness
0.977
0.934
0.914
0.896
0.645
0.180
F1
0.929
0.945
0.903
0.910
0.713
0.304

Overall accuracy 0.908

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9867
0.0036
0.0067
0.0006
0.0023
0.0001
building
0.0357
0.9563
0.0042
0.0003
0.0034
0.0000
low_veg
0.0250
0.0012
0.9429
0.0306
0.0003
0.0000
tree
0.0095
0.0001
0.0512
0.9391
0.0001
0.0000
car
0.0931
0.1907
0.0015
0.0007
0.7139
0.0000
clutter
0.5186
0.1114
0.0706
0.0060
0.0848
0.2086
Precision/Correctness
0.920
0.964
0.924
0.953
0.820
0.996
Recall/Completeness
0.987
0.956
0.943
0.939
0.714
0.209
F1
0.952
0.960
0.933
0.946
0.763
0.345

Overall accuracy 0.935

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9777
0.0112
0.0062
0.0033
0.0014
0.0003
building
0.0822
0.8573
0.0526
0.0041
0.0039
0.0000
low_veg
0.0635
0.0083
0.8828
0.0449
0.0005
0.0000
tree
0.0357
0.0097
0.1166
0.8380
0.0000
0.0000
car
0.3198
0.0138
0.0021
0.0033
0.6611
0.0000
clutter
0.5329
0.0392
0.0008
0.0014
0.0713
0.3544
Precision/Correctness
0.843
0.965
0.820
0.903
0.604
0.994
Recall/Completeness
0.978
0.857
0.883
0.838
0.661
0.354
F1
0.905
0.908
0.850
0.869
0.631
0.523

Overall accuracy 0.876

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9884
0.0082
0.0019
0.0009
0.0006
0.0000
building
0.0735
0.8712
0.0500
0.0017
0.0036
0.0000
low_veg
0.0328
0.0042
0.9212
0.0415
0.0004
0.0000
tree
0.0133
0.0070
0.1040
0.8757
0.0000
0.0000
car
0.2063
0.0120
0.0004
0.0007
0.7805
0.0000
clutter
0.4900
0.0379
0.0001
0.0008
0.0768
0.3944
Precision/Correctness
0.874
0.976
0.841
0.926
0.599
1.000
Recall/Completeness
0.988
0.871
0.921
0.876
0.781
0.394
F1
0.928
0.920
0.879
0.900
0.678
0.566

Overall accuracy 0.900

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8647
0.0509
0.0635
0.0194
0.0015
0.0000
building
0.0371
0.9417
0.0151
0.0054
0.0007
0.0000
low_veg
0.0478
0.0099
0.7061
0.2355
0.0006
0.0000
tree
0.0249
0.0049
0.0863
0.8838
0.0001
0.0000
car
0.2917
0.0045
0.0000
0.0084
0.6955
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.879
0.939
0.814
0.762
0.880
--
Recall/Completeness
0.865
0.942
0.706
0.884
0.695
--
F1
0.872
0.940
0.756
0.818
0.777
--

Overall accuracy 0.849

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9042
0.0481
0.0425
0.0047
0.0005
0.0000
building
0.0269
0.9650
0.0068
0.0007
0.0005
0.0000
low_veg
0.0339
0.0036
0.7362
0.2257
0.0006
0.0000
tree
0.0061
0.0011
0.0618
0.9310
0.0000
0.0000
car
0.1826
0.0058
0.0000
0.0017
0.8099
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.924
0.954
0.872
0.790
0.919
--
Recall/Completeness
0.904
0.965
0.736
0.931
0.810
--
F1
0.914
0.960
0.798
0.855
0.861
--

Overall accuracy 0.886

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9202
0.0181
0.0475
0.0116
0.0026
0.0000
building
0.0335
0.9221
0.0370
0.0057
0.0016
0.0000
low_veg
0.0533
0.0461
0.8212
0.0791
0.0004
0.0000
tree
0.0281
0.0020
0.1098
0.8601
0.0000
0.0000
car
0.3230
0.0043
0.0177
0.0060
0.6491
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.885
0.928
0.840
0.862
0.836
--
Recall/Completeness
0.920
0.922
0.821
0.860
0.649
--
F1
0.902
0.925
0.831
0.861
0.731
--

Overall accuracy 0.879

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


=============================================================

Tile top_mosaic_09cm_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9435
0.0161
0.0351
0.0041
0.0013
0.0000
building
0.0236
0.9510
0.0219
0.0021
0.0014
0.0000
low_veg
0.0342
0.0467
0.8561
0.0628
0.0002
0.0000
tree
0.0104
0.0003
0.0861
0.9032
0.0000
0.0000
car
0.2223
0.0018
0.0113
0.0019
0.7628
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.929
0.936
0.882
0.900
0.879
--
Recall/Completeness
0.943
0.951
0.856
0.903
0.763
--
F1
0.936
0.943
0.869
0.902
0.817
--

Overall accuracy 0.913

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image