Detailled semantic labeling (2D) result


Name X. Sun et al.
Affiliation NLPR, China
Abbreviation NLPR3
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
0.905
0.034
0.041
0.013
0.007
0.000
building
0.034
0.945
0.017
0.004
0.000
0.000
low_veg
0.047
0.023
0.819
0.110
0.000
0.000
tree
0.021
0.006
0.101
0.871
0.000
0.000
car
0.219
0.044
0.007
0.006
0.724
0.000
clutter
0.298
0.288
0.012
0.007
0.023
0.372
Precision/Correctness
0.896
0.928
0.820
0.873
0.804
0.968
Recall/Completeness
0.905
0.945
0.819
0.871
0.724
0.372
F1
0.900
0.936
0.820
0.872
0.762
0.537

Overall accuracy 0.882

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
0.932
0.026
0.032
0.006
0.003
0.000
building
0.027
0.961
0.010
0.001
0.000
0.000
low_veg
0.034
0.013
0.856
0.096
0.000
0.000
tree
0.009
0.002
0.083
0.906
0.000
0.000
car
0.116
0.049
0.004
0.003
0.828
0.000
clutter
0.252
0.305
0.009
0.002
0.024
0.408
Precision/Correctness
0.929
0.950
0.856
0.901
0.862
0.981
Recall/Completeness
0.932
0.961
0.856
0.906
0.828
0.408
F1
0.930
0.956
0.856
0.903
0.845
0.577

Overall accuracy 0.912
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9482
0.0142
0.0221
0.0087
0.0069
0.0000
building
0.0339
0.9456
0.0182
0.0023
0.0000
0.0000
low_veg
0.0665
0.0173
0.7787
0.1374
0.0001
0.0000
tree
0.0313
0.0043
0.1108
0.8517
0.0003
0.0016
car
0.2517
0.0007
0.0007
0.0021
0.7448
0.0000
clutter
0.0000
0.0000
0.0000
0.0105
0.0000
0.9895
Precision/Correctness
0.869
0.961
0.837
0.852
0.865
0.979
Recall/Completeness
0.948
0.946
0.779
0.852
0.745
0.990
F1
0.907
0.953
0.807
0.852
0.800
0.984

Overall accuracy 0.880

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
0.9715
0.0102
0.0116
0.0026
0.0041
0.0000
building
0.0269
0.9631
0.0096
0.0004
0.0000
0.0000
low_veg
0.0463
0.0081
0.8130
0.1325
0.0000
0.0000
tree
0.0142
0.0010
0.0954
0.8884
0.0001
0.0008
car
0.1134
0.0000
0.0000
0.0008
0.8858
0.0000
clutter
0.0000
0.0000
0.0000
0.0018
0.0000
0.9982
Precision/Correctness
0.915
0.980
0.871
0.872
0.909
0.990
Recall/Completeness
0.972
0.963
0.813
0.888
0.886
0.998
F1
0.942
0.971
0.841
0.880
0.897
0.994

Overall accuracy 0.911

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
0.9369
0.0452
0.0095
0.0026
0.0058
0.0000
building
0.0463
0.9485
0.0045
0.0006
0.0001
0.0000
low_veg
0.1432
0.0729
0.5521
0.2307
0.0011
0.0000
tree
0.0579
0.0237
0.0741
0.8425
0.0019
0.0000
car
0.2938
0.0249
0.0001
0.0018
0.6794
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.874
0.967
0.735
0.780
0.839
--
Recall/Completeness
0.937
0.949
0.552
0.842
0.679
--
F1
0.904
0.958
0.630
0.810
0.751
--

Overall accuracy 0.917

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
0.9645
0.0251
0.0073
0.0007
0.0025
0.0000
building
0.0323
0.9640
0.0035
0.0001
0.0001
0.0000
low_veg
0.1029
0.0377
0.6007
0.2583
0.0005
0.0000
tree
0.0263
0.0092
0.0667
0.8971
0.0007
0.0000
car
0.1878
0.0168
0.0000
0.0007
0.7947
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.912
0.984
0.767
0.799
0.917
--
Recall/Completeness
0.964
0.964
0.601
0.897
0.795
--
F1
0.938
0.974
0.674
0.845
0.852
--

Overall accuracy 0.944

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
0.8957
0.0547
0.0354
0.0065
0.0078
0.0000
building
0.0333
0.9524
0.0123
0.0018
0.0001
0.0000
low_veg
0.0826
0.0221
0.7665
0.1284
0.0004
0.0000
tree
0.0155
0.0055
0.0704
0.9084
0.0003
0.0000
car
0.2596
0.0218
0.0002
0.0015
0.7169
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.889
0.945
0.710
0.942
0.846
--
Recall/Completeness
0.896
0.952
0.767
0.908
0.717
--
F1
0.892
0.949
0.737
0.925
0.776
--

Overall accuracy 0.901

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
0.9313
0.0383
0.0256
0.0017
0.0031
0.0000
building
0.0245
0.9678
0.0074
0.0003
0.0000
0.0000
low_veg
0.0659
0.0072
0.8141
0.1128
0.0000
0.0000
tree
0.0046
0.0014
0.0558
0.9382
0.0001
0.0000
car
0.1582
0.0206
0.0000
0.0000
0.8211
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.928
0.968
0.757
0.960
0.927
--
Recall/Completeness
0.931
0.968
0.814
0.938
0.821
--
F1
0.929
0.968
0.784
0.949
0.871
--

Overall accuracy 0.933

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
0.8913
0.0313
0.0664
0.0055
0.0055
0.0000
building
0.0244
0.9603
0.0127
0.0023
0.0004
0.0000
low_veg
0.0422
0.0143
0.9200
0.0230
0.0006
0.0000
tree
0.0271
0.0079
0.1144
0.8505
0.0001
0.0000
car
0.2291
0.0096
0.0145
0.0034
0.7435
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.920
0.938
0.779
0.967
0.867
--
Recall/Completeness
0.891
0.960
0.920
0.851
0.743
--
F1
0.905
0.949
0.844
0.905
0.800
--

Overall accuracy 0.901

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
0.9185
0.0232
0.0549
0.0011
0.0023
0.0000
building
0.0150
0.9789
0.0050
0.0007
0.0004
0.0000
low_veg
0.0266
0.0058
0.9521
0.0154
0.0001
0.0000
tree
0.0107
0.0035
0.0895
0.8963
0.0000
0.0000
car
0.1222
0.0079
0.0073
0.0018
0.8609
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.957
0.962
0.822
0.984
0.930
--
Recall/Completeness
0.919
0.979
0.952
0.896
0.861
--
F1
0.937
0.970
0.882
0.938
0.894
--

Overall accuracy 0.933

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
0.8503
0.0746
0.0554
0.0151
0.0045
0.0000
building
0.0264
0.9502
0.0138
0.0092
0.0003
0.0000
low_veg
0.0660
0.0319
0.7649
0.1371
0.0002
0.0000
tree
0.0156
0.0085
0.0884
0.8873
0.0001
0.0000
car
0.2299
0.0206
0.0008
0.0024
0.7464
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.934
0.863
0.761
0.851
0.743
--
Recall/Completeness
0.850
0.950
0.765
0.887
0.746
--
F1
0.890
0.905
0.763
0.869
0.745
--

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
0.8809
0.0619
0.0478
0.0060
0.0034
0.0000
building
0.0206
0.9636
0.0090
0.0066
0.0002
0.0000
low_veg
0.0628
0.0170
0.8094
0.1106
0.0001
0.0000
tree
0.0054
0.0037
0.0667
0.9242
0.0000
0.0000
car
0.1216
0.0284
0.0000
0.0002
0.8498
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.951
0.894
0.801
0.891
0.762
--
Recall/Completeness
0.881
0.964
0.809
0.924
0.850
--
F1
0.915
0.928
0.805
0.907
0.804
--

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
0.8746
0.0426
0.0559
0.0246
0.0023
0.0000
building
0.0732
0.9113
0.0092
0.0063
0.0000
0.0000
low_veg
0.0248
0.0263
0.8111
0.1367
0.0001
0.0010
tree
0.0084
0.0068
0.1010
0.8838
0.0000
0.0000
car
0.2330
0.0011
0.0021
0.0048
0.7590
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.830
0.915
0.833
0.881
0.914
--
Recall/Completeness
0.875
0.911
0.811
0.884
0.759
--
F1
0.852
0.913
0.822
0.882
0.829
--

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_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9033
0.0393
0.0452
0.0116
0.0007
0.0000
building
0.0750
0.9200
0.0030
0.0019
0.0000
0.0000
low_veg
0.0223
0.0149
0.8484
0.1132
0.0000
0.0011
tree
0.0035
0.0024
0.0817
0.9124
0.0000
0.0000
car
0.0834
0.0001
0.0002
0.0004
0.9159
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.851
0.943
0.865
0.909
0.969
--
Recall/Completeness
0.903
0.920
0.848
0.912
0.916
--
F1
0.877
0.932
0.857
0.911
0.942
--

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
0.8427
0.0295
0.0953
0.0262
0.0063
0.0000
building
0.0054
0.9777
0.0130
0.0039
0.0000
0.0000
low_veg
0.0758
0.0534
0.7491
0.1215
0.0002
0.0000
tree
0.0249
0.0142
0.1299
0.8310
0.0000
0.0000
car
0.1609
0.0000
0.0000
0.0085
0.8305
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.870
0.873
0.738
0.881
0.599
--
Recall/Completeness
0.843
0.978
0.749
0.831
0.831
--
F1
0.856
0.922
0.743
0.855
0.696
--

Overall accuracy 0.839

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
0.8617
0.0241
0.0916
0.0170
0.0056
0.0000
building
0.0013
0.9922
0.0056
0.0009
0.0000
0.0000
low_veg
0.0706
0.0362
0.7822
0.1110
0.0001
0.0000
tree
0.0166
0.0067
0.1163
0.8605
0.0000
0.0000
car
0.0442
0.0000
0.0000
0.0037
0.9522
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.895
0.911
0.759
0.902
0.584
--
Recall/Completeness
0.862
0.992
0.782
0.860
0.952
--
F1
0.878
0.950
0.771
0.881
0.724
--

Overall accuracy 0.864

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
0.8033
0.0403
0.1049
0.0479
0.0037
0.0000
building
0.0162
0.9477
0.0256
0.0104
0.0001
0.0000
low_veg
0.0107
0.0157
0.8778
0.0958
0.0000
0.0000
tree
0.0196
0.0061
0.1141
0.8602
0.0000
0.0000
car
0.1694
0.0047
0.0024
0.0078
0.8157
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.901
0.928
0.806
0.893
0.753
--
Recall/Completeness
0.803
0.948
0.878
0.860
0.816
--
F1
0.849
0.938
0.840
0.876
0.783
--

Overall accuracy 0.872

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
0.8377
0.0354
0.0964
0.0288
0.0018
0.0000
building
0.0129
0.9689
0.0144
0.0037
0.0000
0.0000
low_veg
0.0075
0.0090
0.9024
0.0811
0.0000
0.0000
tree
0.0124
0.0020
0.0985
0.8870
0.0000
0.0000
car
0.0970
0.0011
0.0001
0.0015
0.9002
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.932
0.952
0.834
0.918
0.838
--
Recall/Completeness
0.838
0.969
0.902
0.887
0.900
--
F1
0.882
0.960
0.867
0.902
0.868
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8700
0.0382
0.0425
0.0439
0.0054
0.0000
building
0.0143
0.9580
0.0223
0.0054
0.0000
0.0000
low_veg
0.0390
0.0387
0.7807
0.1415
0.0000
0.0000
tree
0.0084
0.0024
0.0535
0.9357
0.0000
0.0000
car
0.1193
0.0084
0.0214
0.0151
0.8358
0.0000
clutter
0.0024
0.9964
0.0002
0.0010
0.0000
0.0000
Precision/Correctness
0.926
0.784
0.815
0.912
0.759
-nan
Recall/Completeness
0.870
0.958
0.781
0.936
0.836
0.000
F1
0.897
0.862
0.798
0.924
0.795
-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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9032
0.0275
0.0399
0.0261
0.0034
0.0000
building
0.0120
0.9674
0.0194
0.0012
0.0000
0.0000
low_veg
0.0325
0.0258
0.8285
0.1133
0.0000
0.0000
tree
0.0034
0.0004
0.0406
0.9555
0.0000
0.0000
car
0.0309
0.0047
0.0087
0.0051
0.9506
0.0000
clutter
0.0005
0.9995
0.0000
0.0000
0.0000
0.0000
Precision/Correctness
0.950
0.806
0.844
0.938
0.807
-nan
Recall/Completeness
0.903
0.967
0.828
0.955
0.951
0.000
F1
0.926
0.880
0.836
0.947
0.873
-nan

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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8984
0.0281
0.0381
0.0277
0.0078
0.0000
building
0.0256
0.9604
0.0080
0.0058
0.0001
0.0000
low_veg
0.0734
0.0438
0.7638
0.1185
0.0004
0.0000
tree
0.0214
0.0046
0.0657
0.9082
0.0002
0.0000
car
0.2219
0.0000
0.0105
0.0198
0.7478
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.911
0.899
0.825
0.885
0.757
--
Recall/Completeness
0.898
0.960
0.764
0.908
0.748
--
F1
0.905
0.929
0.793
0.896
0.752
--

Overall accuracy 0.883

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
0.9299
0.0200
0.0318
0.0131
0.0052
0.0000
building
0.0194
0.9752
0.0028
0.0025
0.0000
0.0000
low_veg
0.0695
0.0282
0.8111
0.0908
0.0003
0.0000
tree
0.0083
0.0012
0.0479
0.9425
0.0001
0.0000
car
0.1091
0.0000
0.0086
0.0088
0.8736
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.937
0.935
0.865
0.923
0.799
--
Recall/Completeness
0.930
0.975
0.811
0.942
0.874
--
F1
0.933
0.955
0.837
0.933
0.835
--

Overall accuracy 0.918

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
0.8212
0.0377
0.1015
0.0322
0.0073
0.0000
building
0.0576
0.8991
0.0358
0.0075
0.0000
0.0000
low_veg
0.0135
0.0193
0.9113
0.0557
0.0002
0.0000
tree
0.0162
0.0072
0.1545
0.8221
0.0001
0.0000
car
0.1717
0.0000
0.0042
0.0017
0.8223
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.850
0.917
0.861
0.878
0.730
--
Recall/Completeness
0.821
0.899
0.911
0.822
0.822
--
F1
0.835
0.908
0.885
0.849
0.773
--

Overall accuracy 0.873

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
0.8537
0.0271
0.0918
0.0228
0.0046
0.0000
building
0.0554
0.9112
0.0287
0.0046
0.0000
0.0000
low_veg
0.0102
0.0140
0.9314
0.0442
0.0002
0.0000
tree
0.0098
0.0028
0.1316
0.8558
0.0000
0.0000
car
0.0783
0.0000
0.0001
0.0000
0.9216
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.875
0.943
0.883
0.904
0.772
--
Recall/Completeness
0.854
0.911
0.931
0.856
0.922
--
F1
0.864
0.927
0.907
0.879
0.840
--

Overall accuracy 0.898

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
0.9332
0.0308
0.0186
0.0053
0.0121
0.0000
building
0.0543
0.9305
0.0126
0.0025
0.0002
0.0000
low_veg
0.0822
0.0296
0.7255
0.1619
0.0008
0.0000
tree
0.0150
0.0071
0.0455
0.9323
0.0001
0.0000
car
0.2666
0.0140
0.0067
0.0051
0.7075
0.0000
clutter
0.4679
0.2880
0.1412
0.0031
0.0997
0.0000
Precision/Correctness
0.865
0.949
0.902
0.731
0.841
-nan
Recall/Completeness
0.933
0.930
0.726
0.932
0.708
0.000
F1
0.898
0.940
0.804
0.819
0.768
-nan

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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9614
0.0202
0.0118
0.0019
0.0048
0.0000
building
0.0436
0.9482
0.0074
0.0007
0.0001
0.0000
low_veg
0.0618
0.0148
0.7726
0.1505
0.0003
0.0000
tree
0.0062
0.0021
0.0300
0.9617
0.0000
0.0000
car
0.1592
0.0135
0.0052
0.0040
0.8181
0.0000
clutter
0.4783
0.2712
0.1257
0.0008
0.1240
0.0000
Precision/Correctness
0.902
0.971
0.938
0.761
0.915
-nan
Recall/Completeness
0.961
0.948
0.773
0.962
0.818
0.000
F1
0.931
0.959
0.847
0.850
0.864
-nan

Overall accuracy 0.915

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
0.9393
0.0185
0.0245
0.0111
0.0067
0.0000
building
0.0208
0.9631
0.0101
0.0060
0.0000
0.0000
low_veg
0.0563
0.0222
0.6666
0.2538
0.0011
0.0000
tree
0.0055
0.0032
0.0238
0.9676
0.0000
0.0000
car
0.2047
0.0053
0.0076
0.0124
0.7700
0.0000
clutter
0.3902
0.4131
0.0262
0.0074
0.0375
0.1256
Precision/Correctness
0.932
0.917
0.856
0.705
0.716
0.995
Recall/Completeness
0.939
0.963
0.667
0.968
0.770
0.126
F1
0.936
0.939
0.750
0.816
0.742
0.223

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9613
0.0118
0.0174
0.0050
0.0045
0.0000
building
0.0147
0.9773
0.0050
0.0029
0.0000
0.0000
low_veg
0.0332
0.0084
0.7164
0.2413
0.0007
0.0000
tree
0.0012
0.0006
0.0156
0.9826
0.0000
0.0000
car
0.1052
0.0039
0.0085
0.0096
0.8728
0.0000
clutter
0.3476
0.4432
0.0183
0.0017
0.0402
0.1490
Precision/Correctness
0.956
0.938
0.898
0.743
0.742
0.998
Recall/Completeness
0.961
0.977
0.716
0.983
0.873
0.149
F1
0.958
0.957
0.797
0.846
0.802
0.259

Overall accuracy 0.911

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
0.9630
0.0120
0.0086
0.0036
0.0123
0.0005
building
0.0298
0.9596
0.0051
0.0026
0.0027
0.0002
low_veg
0.0435
0.0064
0.9185
0.0294
0.0008
0.0015
tree
0.0365
0.0034
0.0716
0.8867
0.0019
0.0000
car
0.1190
0.1944
0.0041
0.0055
0.6770
0.0000
clutter
0.4457
0.2273
0.0179
0.0147
0.0453
0.2491
Precision/Correctness
0.908
0.928
0.910
0.933
0.767
0.947
Recall/Completeness
0.963
0.960
0.918
0.887
0.677
0.249
F1
0.934
0.943
0.914
0.909
0.719
0.394

Overall accuracy 0.912

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
0.9800
0.0081
0.0057
0.0007
0.0053
0.0002
building
0.0180
0.9784
0.0010
0.0003
0.0021
0.0002
low_veg
0.0247
0.0031
0.9488
0.0230
0.0001
0.0003
tree
0.0107
0.0004
0.0535
0.9350
0.0005
0.0000
car
0.0506
0.2168
0.0007
0.0013
0.7306
0.0000
clutter
0.3916
0.2507
0.0173
0.0062
0.0504
0.2838
Precision/Correctness
0.942
0.942
0.941
0.962
0.831
0.980
Recall/Completeness
0.980
0.978
0.949
0.935
0.731
0.284
F1
0.961
0.960
0.945
0.948
0.777
0.440

Overall accuracy 0.942

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
0.9720
0.0163
0.0065
0.0015
0.0026
0.0010
building
0.0520
0.8899
0.0538
0.0028
0.0010
0.0006
low_veg
0.0600
0.0096
0.9118
0.0184
0.0002
0.0000
tree
0.0506
0.0054
0.1477
0.7962
0.0000
0.0000
car
0.2319
0.0561
0.0019
0.0026
0.7076
0.0000
clutter
0.4366
0.0374
0.0025
0.0020
0.0147
0.5069
Precision/Correctness
0.869
0.959
0.806
0.950
0.811
0.980
Recall/Completeness
0.972
0.890
0.912
0.796
0.708
0.507
F1
0.918
0.923
0.856
0.867
0.756
0.668

Overall accuracy 0.892

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
0.9834
0.0128
0.0020
0.0002
0.0012
0.0003
building
0.0435
0.9037
0.0504
0.0010
0.0008
0.0005
low_veg
0.0296
0.0058
0.9484
0.0161
0.0001
0.0000
tree
0.0218
0.0010
0.1343
0.8429
0.0000
0.0000
car
0.1216
0.0508
0.0000
0.0011
0.8265
0.0000
clutter
0.3853
0.0326
0.0012
0.0010
0.0165
0.5634
Precision/Correctness
0.904
0.971
0.828
0.968
0.841
0.991
Recall/Completeness
0.983
0.904
0.948
0.843
0.827
0.563
F1
0.942
0.936
0.884
0.901
0.834
0.718

Overall accuracy 0.917

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
0.8690
0.0488
0.0692
0.0107
0.0023
0.0000
building
0.0250
0.9575
0.0144
0.0026
0.0005
0.0000
low_veg
0.0461
0.0120
0.7695
0.1723
0.0001
0.0000
tree
0.0344
0.0074
0.1363
0.8216
0.0002
0.0000
car
0.2647
0.0029
0.0000
0.0085
0.7239
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.886
0.938
0.782
0.806
0.874
--
Recall/Completeness
0.869
0.958
0.769
0.822
0.724
--
F1
0.877
0.947
0.776
0.814
0.792
--

Overall accuracy 0.856

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
0.9081
0.0437
0.0460
0.0018
0.0005
0.0000
building
0.0155
0.9789
0.0048
0.0003
0.0005
0.0000
low_veg
0.0295
0.0043
0.7968
0.1693
0.0000
0.0000
tree
0.0100
0.0023
0.1056
0.8820
0.0000
0.0000
car
0.1466
0.0014
0.0000
0.0032
0.8488
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.938
0.957
0.841
0.827
0.950
--
Recall/Completeness
0.908
0.979
0.797
0.882
0.849
--
F1
0.923
0.968
0.818
0.854
0.896
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8987
0.0359
0.0485
0.0113
0.0053
0.0004
building
0.0205
0.9440
0.0308
0.0031
0.0016
0.0000
low_veg
0.0315
0.0135
0.8858
0.0686
0.0005
0.0000
tree
0.0221
0.0043
0.1345
0.8390
0.0001
0.0000
car
0.1867
0.0380
0.0373
0.0084
0.7296
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.923
0.942
0.840
0.876
0.780
--
Recall/Completeness
0.899
0.944
0.886
0.839
0.730
--
F1
0.911
0.943
0.862
0.857
0.754
--

Overall accuracy 0.893

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
0.9275
0.0330
0.0332
0.0034
0.0024
0.0004
building
0.0137
0.9684
0.0157
0.0009
0.0013
0.0000
low_veg
0.0164
0.0075
0.9211
0.0547
0.0003
0.0000
tree
0.0069
0.0010
0.1091
0.8830
0.0000
0.0000
car
0.0971
0.0446
0.0230
0.0024
0.8330
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.961
0.957
0.882
0.911
0.848
--
Recall/Completeness
0.928
0.968
0.921
0.883
0.833
--
F1
0.944
0.963
0.901
0.897
0.840
--

Overall accuracy 0.927

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image