Detailled semantic labeling (2D) result


Name S. Ghassemi
Affiliation Politecnico di Torino, Department of Electronic and Telecommunication , Turin, Italy
Abbreviation TOR_1
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
0.888
0.038
0.049
0.017
0.006
0.003
building
0.034
0.934
0.026
0.006
0.000
0.000
low_veg
0.074
0.026
0.757
0.142
0.001
0.001
tree
0.023
0.007
0.107
0.862
0.000
0.001
car
0.256
0.061
0.011
0.006
0.664
0.001
clutter
0.221
0.312
0.035
0.097
0.018
0.317
Precision/Correctness
0.875
0.919
0.785
0.838
0.803
0.693
Recall/Completeness
0.888
0.934
0.757
0.862
0.664
0.317
F1
0.881
0.926
0.771
0.850
0.727
0.435

Overall accuracy 0.858

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
0.916
0.031
0.038
0.008
0.004
0.003
building
0.027
0.952
0.018
0.003
0.000
0.000
low_veg
0.065
0.018
0.788
0.128
0.001
0.001
tree
0.012
0.003
0.089
0.894
0.000
0.001
car
0.166
0.069
0.007
0.004
0.753
0.001
clutter
0.180
0.331
0.031
0.097
0.018
0.342
Precision/Correctness
0.904
0.939
0.823
0.867
0.841
0.716
Recall/Completeness
0.916
0.952
0.788
0.894
0.753
0.342
F1
0.910
0.945
0.805
0.880
0.795
0.463

Overall accuracy 0.887
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9408
0.0081
0.0319
0.0104
0.0087
0.0001
building
0.0207
0.9451
0.0303
0.0038
0.0001
0.0000
low_veg
0.0660
0.0209
0.7648
0.1479
0.0004
0.0000
tree
0.0288
0.0038
0.1070
0.8583
0.0003
0.0017
car
0.2337
0.0009
0.0018
0.0009
0.7623
0.0003
clutter
0.0000
0.0000
0.0000
0.0153
0.0000
0.9847
Precision/Correctness
0.882
0.964
0.819
0.841
0.835
0.975
Recall/Completeness
0.941
0.945
0.765
0.858
0.762
0.985
F1
0.910
0.954
0.791
0.850
0.797
0.980

Overall accuracy 0.877

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
0.9672
0.0048
0.0187
0.0030
0.0061
0.0001
building
0.0135
0.9680
0.0171
0.0014
0.0000
0.0000
low_veg
0.0498
0.0144
0.7938
0.1418
0.0003
0.0000
tree
0.0139
0.0019
0.0907
0.8925
0.0001
0.0009
car
0.1250
0.0000
0.0001
0.0002
0.8745
0.0002
clutter
0.0000
0.0000
0.0000
0.0058
0.0000
0.9942
Precision/Correctness
0.922
0.978
0.859
0.864
0.865
0.987
Recall/Completeness
0.967
0.968
0.794
0.892
0.875
0.994
F1
0.944
0.973
0.825
0.878
0.870
0.991

Overall accuracy 0.907

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
0.8775
0.0496
0.0614
0.0089
0.0026
0.0000
building
0.0666
0.8983
0.0288
0.0058
0.0004
0.0000
low_veg
0.0947
0.0231
0.5896
0.2926
0.0000
0.0000
tree
0.0298
0.0049
0.0960
0.8694
0.0000
0.0000
car
0.5031
0.0037
0.0266
0.0057
0.4609
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.836
0.970
0.402
0.687
0.868
--
Recall/Completeness
0.877
0.898
0.590
0.869
0.461
--
F1
0.856
0.933
0.478
0.767
0.602
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9089
0.0324
0.0526
0.0046
0.0016
0.0000
building
0.0556
0.9157
0.0244
0.0037
0.0005
0.0000
low_veg
0.0701
0.0060
0.6145
0.3094
0.0000
0.0000
tree
0.0099
0.0006
0.0778
0.9117
0.0000
0.0000
car
0.4160
0.0006
0.0234
0.0038
0.5563
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.865
0.983
0.425
0.729
0.895
--
Recall/Completeness
0.909
0.916
0.614
0.912
0.556
--
F1
0.886
0.948
0.503
0.810
0.686
--

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_area6_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8994
0.0386
0.0462
0.0117
0.0040
0.0000
building
0.0516
0.9233
0.0203
0.0047
0.0001
0.0000
low_veg
0.0788
0.0188
0.7651
0.1369
0.0004
0.0000
tree
0.0144
0.0030
0.0607
0.9218
0.0001
0.0000
car
0.3561
0.0238
0.0069
0.0013
0.6120
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.866
0.959
0.693
0.933
0.898
--
Recall/Completeness
0.899
0.923
0.765
0.922
0.612
--
F1
0.882
0.941
0.727
0.927
0.728
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9334
0.0249
0.0357
0.0045
0.0015
0.0000
building
0.0415
0.9430
0.0136
0.0018
0.0001
0.0000
low_veg
0.0644
0.0099
0.8085
0.1171
0.0000
0.0000
tree
0.0049
0.0008
0.0461
0.9481
0.0000
0.0000
car
0.2635
0.0227
0.0042
0.0004
0.7092
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.902
0.977
0.742
0.955
0.954
--
Recall/Completeness
0.933
0.943
0.809
0.948
0.709
--
F1
0.918
0.960
0.774
0.952
0.814
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8765
0.0320
0.0698
0.0139
0.0068
0.0010
building
0.0273
0.9530
0.0155
0.0030
0.0005
0.0006
low_veg
0.0518
0.0156
0.8733
0.0588
0.0005
0.0000
tree
0.0174
0.0053
0.0927
0.8840
0.0007
0.0000
car
0.2026
0.0050
0.0211
0.0065
0.7647
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.919
0.938
0.781
0.927
0.838
--
Recall/Completeness
0.876
0.953
0.873
0.884
0.765
--
F1
0.897
0.946
0.825
0.905
0.800
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9084
0.0245
0.0574
0.0051
0.0036
0.0010
building
0.0198
0.9716
0.0070
0.0007
0.0004
0.0005
low_veg
0.0368
0.0079
0.9094
0.0457
0.0002
0.0000
tree
0.0057
0.0018
0.0718
0.9201
0.0006
0.0000
car
0.1144
0.0023
0.0145
0.0052
0.8635
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.951
0.960
0.827
0.954
0.886
--
Recall/Completeness
0.908
0.972
0.909
0.920
0.864
--
F1
0.929
0.966
0.866
0.937
0.875
--

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
0.7981
0.1170
0.0559
0.0215
0.0075
0.0000
building
0.0204
0.9488
0.0155
0.0151
0.0002
0.0000
low_veg
0.0874
0.0254
0.7231
0.1633
0.0009
0.0000
tree
0.0149
0.0074
0.0879
0.8894
0.0004
0.0000
car
0.1660
0.0313
0.0085
0.0012
0.7930
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.925
0.817
0.748
0.819
0.644
--
Recall/Completeness
0.798
0.949
0.723
0.889
0.793
--
F1
0.857
0.878
0.735
0.853
0.711
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8287
0.1085
0.0469
0.0101
0.0057
0.0000
building
0.0148
0.9628
0.0106
0.0116
0.0002
0.0000
low_veg
0.0865
0.0127
0.7669
0.1332
0.0007
0.0000
tree
0.0061
0.0032
0.0698
0.9206
0.0003
0.0000
car
0.0928
0.0343
0.0080
0.0001
0.8648
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.941
0.839
0.789
0.863
0.652
--
Recall/Completeness
0.829
0.963
0.767
0.921
0.865
--
F1
0.881
0.897
0.778
0.891
0.744
--

Overall accuracy 0.869

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
0.8761
0.0574
0.0404
0.0178
0.0052
0.0031
building
0.0190
0.9620
0.0062
0.0033
0.0000
0.0094
low_veg
0.1213
0.0445
0.7065
0.1259
0.0009
0.0009
tree
0.0209
0.0119
0.1352
0.8307
0.0001
0.0013
car
0.1541
0.0033
0.0000
0.0008
0.8418
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.725
0.878
0.782
0.886
0.807
--
Recall/Completeness
0.876
0.962
0.707
0.831
0.842
--
F1
0.793
0.918
0.743
0.857
0.824
--

Overall accuracy 0.826

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
0.9021
0.0539
0.0307
0.0089
0.0020
0.0024
building
0.0168
0.9697
0.0024
0.0009
0.0000
0.0102
low_veg
0.1228
0.0317
0.7382
0.1056
0.0008
0.0009
tree
0.0138
0.0055
0.1175
0.8621
0.0000
0.0010
car
0.0526
0.0002
0.0000
0.0000
0.9472
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.748
0.910
0.812
0.911
0.870
--
Recall/Completeness
0.902
0.970
0.738
0.862
0.947
--
F1
0.818
0.939
0.773
0.886
0.907
--

Overall accuracy 0.854

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
0.8268
0.0828
0.0500
0.0330
0.0073
0.0001
building
0.0106
0.9685
0.0159
0.0050
0.0000
0.0000
low_veg
0.0950
0.0515
0.7170
0.1349
0.0016
0.0000
tree
0.0295
0.0139
0.1210
0.8355
0.0001
0.0000
car
0.1120
0.0000
0.0000
0.0045
0.8835
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.840
0.822
0.769
0.867
0.526
--
Recall/Completeness
0.827
0.969
0.717
0.836
0.884
--
F1
0.833
0.890
0.742
0.851
0.660
--

Overall accuracy 0.827

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
0.8472
0.0791
0.0445
0.0227
0.0063
0.0001
building
0.0065
0.9839
0.0076
0.0020
0.0000
0.0000
low_veg
0.0906
0.0348
0.7494
0.1235
0.0016
0.0000
tree
0.0213
0.0070
0.1083
0.8635
0.0000
0.0000
car
0.0322
0.0000
0.0000
0.0010
0.9668
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.865
0.853
0.794
0.889
0.502
--
Recall/Completeness
0.847
0.984
0.749
0.863
0.967
--
F1
0.856
0.914
0.771
0.876
0.661
--

Overall accuracy 0.852

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
0.8507
0.0594
0.0423
0.0402
0.0067
0.0007
building
0.0121
0.9733
0.0088
0.0056
0.0001
0.0001
low_veg
0.0966
0.0427
0.7685
0.0887
0.0010
0.0026
tree
0.0338
0.0116
0.1337
0.8203
0.0001
0.0005
car
0.0806
0.0027
0.0008
0.0000
0.9158
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.752
0.869
0.799
0.900
0.597
--
Recall/Completeness
0.851
0.973
0.768
0.820
0.916
--
F1
0.798
0.918
0.784
0.858
0.723
--

Overall accuracy 0.837

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
0.8840
0.0511
0.0357
0.0242
0.0044
0.0006
building
0.0094
0.9850
0.0034
0.0021
0.0000
0.0000
low_veg
0.0953
0.0330
0.7924
0.0757
0.0010
0.0027
tree
0.0247
0.0056
0.1207
0.8486
0.0000
0.0004
car
0.0265
0.0000
0.0000
0.0000
0.9735
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.773
0.900
0.824
0.922
0.613
--
Recall/Completeness
0.884
0.985
0.792
0.849
0.974
--
F1
0.825
0.941
0.808
0.884
0.752
--

Overall accuracy 0.862

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
0.8556
0.0713
0.0312
0.0299
0.0067
0.0052
building
0.0158
0.9749
0.0055
0.0038
0.0000
0.0000
low_veg
0.1498
0.0351
0.6847
0.1281
0.0011
0.0012
tree
0.0137
0.0067
0.0798
0.8996
0.0000
0.0002
car
0.1095
0.0156
0.0095
0.0019
0.8635
0.0000
clutter
0.1009
0.8981
0.0008
0.0002
0.0000
0.0000
Precision/Correctness
0.810
0.768
0.768
0.922
0.695
0.000
Recall/Completeness
0.856
0.975
0.685
0.900
0.864
0.000
F1
0.832
0.859
0.724
0.910
0.770
-nan

Overall accuracy 0.837

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
0.8868
0.0605
0.0249
0.0175
0.0048
0.0054
building
0.0130
0.9835
0.0024
0.0010
0.0000
0.0000
low_veg
0.1491
0.0216
0.7248
0.1023
0.0011
0.0011
tree
0.0070
0.0045
0.0658
0.9226
0.0000
0.0001
car
0.0410
0.0135
0.0009
0.0000
0.9446
0.0000
clutter
0.0935
0.9064
0.0001
0.0000
0.0000
0.0000
Precision/Correctness
0.834
0.790
0.803
0.945
0.705
0.000
Recall/Completeness
0.887
0.984
0.725
0.923
0.945
0.000
F1
0.860
0.876
0.762
0.933
0.807
-nan

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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8899
0.0246
0.0444
0.0308
0.0101
0.0002
building
0.0237
0.9512
0.0183
0.0066
0.0002
0.0000
low_veg
0.1041
0.0383
0.7118
0.1450
0.0009
0.0000
tree
0.0218
0.0036
0.0704
0.9040
0.0003
0.0000
car
0.1863
0.0065
0.0086
0.0109
0.7877
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.894
0.910
0.792
0.866
0.711
--
Recall/Completeness
0.890
0.951
0.712
0.904
0.788
--
F1
0.892
0.930
0.750
0.884
0.747
--

Overall accuracy 0.868

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
0.9240
0.0164
0.0366
0.0160
0.0068
0.0002
building
0.0192
0.9677
0.0100
0.0030
0.0001
0.0000
low_veg
0.1030
0.0269
0.7533
0.1160
0.0007
0.0000
tree
0.0093
0.0009
0.0532
0.9366
0.0001
0.0000
car
0.1004
0.0077
0.0059
0.0041
0.8819
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.917
0.942
0.835
0.904
0.751
--
Recall/Completeness
0.924
0.968
0.753
0.937
0.882
--
F1
0.921
0.955
0.792
0.920
0.811
--

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_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7964
0.0531
0.0581
0.0249
0.0142
0.0534
building
0.0438
0.9350
0.0171
0.0041
0.0000
0.0000
low_veg
0.0569
0.0279
0.8679
0.0457
0.0008
0.0009
tree
0.0267
0.0110
0.1574
0.7976
0.0001
0.0072
car
0.1141
0.0027
0.0007
0.0017
0.8389
0.0419
clutter
--
--
--
--
--
--
Precision/Correctness
0.749
0.889
0.874
0.897
0.568
--
Recall/Completeness
0.796
0.935
0.868
0.798
0.839
--
F1
0.772
0.911
0.871
0.845
0.678
--

Overall accuracy 0.852

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
0.8249
0.0400
0.0519
0.0171
0.0115
0.0546
building
0.0418
0.9429
0.0129
0.0023
0.0000
0.0000
low_veg
0.0531
0.0212
0.8905
0.0336
0.0008
0.0009
tree
0.0191
0.0065
0.1389
0.8287
0.0001
0.0067
car
0.0628
0.0040
0.0000
0.0000
0.8963
0.0369
clutter
--
--
--
--
--
--
Precision/Correctness
0.771
0.917
0.893
0.924
0.551
--
Recall/Completeness
0.825
0.943
0.890
0.829
0.896
--
F1
0.797
0.930
0.892
0.874
0.682
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9197
0.0171
0.0451
0.0095
0.0085
0.0000
building
0.0673
0.8860
0.0392
0.0074
0.0001
0.0000
low_veg
0.0532
0.0154
0.7074
0.2239
0.0001
0.0000
tree
0.0074
0.0024
0.0341
0.9560
0.0000
0.0000
car
0.2914
0.0123
0.0126
0.0075
0.6762
0.0000
clutter
0.6891
0.0174
0.2153
0.0097
0.0685
0.0000
Precision/Correctness
0.865
0.971
0.810
0.658
0.881
0.000
Recall/Completeness
0.920
0.886
0.707
0.956
0.676
0.000
F1
0.892
0.927
0.755
0.780
0.765
-nan

Overall accuracy 0.860

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
0.9496
0.0098
0.0340
0.0040
0.0027
0.0000
building
0.0578
0.9094
0.0292
0.0035
0.0001
0.0000
low_veg
0.0408
0.0066
0.7429
0.2096
0.0000
0.0000
tree
0.0021
0.0004
0.0210
0.9764
0.0000
0.0000
car
0.1956
0.0137
0.0090
0.0055
0.7762
0.0000
clutter
0.7186
0.0107
0.1905
0.0078
0.0723
0.0000
Precision/Correctness
0.896
0.986
0.854
0.692
0.947
0.000
Recall/Completeness
0.950
0.909
0.743
0.976
0.776
0.000
F1
0.922
0.946
0.794
0.810
0.853
-nan

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9139
0.0121
0.0522
0.0176
0.0038
0.0005
building
0.0252
0.9273
0.0364
0.0110
0.0001
0.0000
low_veg
0.0525
0.0112
0.6093
0.3269
0.0001
0.0001
tree
0.0066
0.0013
0.0247
0.9674
0.0000
0.0000
car
0.2757
0.0043
0.0097
0.0108
0.6987
0.0008
clutter
0.2749
0.4571
0.0795
0.0395
0.0253
0.1236
Precision/Correctness
0.934
0.925
0.708
0.638
0.798
0.941
Recall/Completeness
0.914
0.927
0.609
0.967
0.699
0.124
F1
0.924
0.926
0.655
0.769
0.745
0.218

Overall accuracy 0.852

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
0.9401
0.0063
0.0415
0.0093
0.0024
0.0005
building
0.0218
0.9445
0.0273
0.0063
0.0001
0.0000
low_veg
0.0397
0.0035
0.6433
0.3134
0.0000
0.0001
tree
0.0025
0.0002
0.0163
0.9811
0.0000
0.0000
car
0.1751
0.0011
0.0095
0.0075
0.8057
0.0011
clutter
0.2332
0.4869
0.0790
0.0289
0.0271
0.1449
Precision/Correctness
0.953
0.942
0.749
0.678
0.820
0.948
Recall/Completeness
0.940
0.945
0.643
0.981
0.806
0.145
F1
0.947
0.943
0.692
0.802
0.813
0.251

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9147
0.0172
0.0565
0.0072
0.0029
0.0015
building
0.0326
0.9407
0.0194
0.0062
0.0006
0.0007
low_veg
0.0269
0.0030
0.9164
0.0487
0.0000
0.0049
tree
0.0340
0.0031
0.0765
0.8847
0.0006
0.0011
car
0.2141
0.3234
0.0119
0.0109
0.4394
0.0002
clutter
0.2614
0.2575
0.0626
0.2901
0.0349
0.0935
Precision/Correctness
0.914
0.903
0.780
0.821
0.862
0.657
Recall/Completeness
0.915
0.941
0.916
0.885
0.439
0.093
F1
0.914
0.922
0.843
0.852
0.582
0.164

Overall accuracy 0.874

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
0.9328
0.0100
0.0526
0.0024
0.0012
0.0009
building
0.0254
0.9594
0.0113
0.0028
0.0004
0.0007
low_veg
0.0130
0.0015
0.9416
0.0393
0.0000
0.0047
tree
0.0123
0.0005
0.0617
0.9247
0.0002
0.0005
car
0.1327
0.3535
0.0059
0.0061
0.5017
0.0002
clutter
0.1988
0.2846
0.0597
0.3190
0.0380
0.0998
Precision/Correctness
0.947
0.924
0.812
0.853
0.890
0.696
Recall/Completeness
0.933
0.959
0.942
0.925
0.502
0.100
F1
0.940
0.941
0.872
0.887
0.642
0.175

Overall accuracy 0.903

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
0.9666
0.0068
0.0140
0.0059
0.0013
0.0054
building
0.0194
0.9217
0.0533
0.0056
0.0000
0.0000
low_veg
0.0613
0.0077
0.4495
0.4805
0.0002
0.0009
tree
0.0357
0.0071
0.1305
0.8267
0.0000
0.0000
car
0.3891
0.0061
0.0105
0.0043
0.5898
0.0001
clutter
0.3100
0.1418
0.0092
0.0893
0.0155
0.4343
Precision/Correctness
0.902
0.956
0.663
0.512
0.846
0.913
Recall/Completeness
0.967
0.922
0.449
0.827
0.590
0.434
F1
0.933
0.939
0.536
0.632
0.695
0.589

Overall accuracy 0.826

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
0.9845
0.0027
0.0079
0.0020
0.0006
0.0023
building
0.0129
0.9346
0.0492
0.0033
0.0000
0.0000
low_veg
0.0368
0.0038
0.4423
0.5160
0.0002
0.0010
tree
0.0154
0.0029
0.1213
0.8603
0.0000
0.0000
car
0.2876
0.0043
0.0079
0.0019
0.6982
0.0001
clutter
0.2671
0.1475
0.0068
0.0981
0.0175
0.4630
Precision/Correctness
0.933
0.968
0.685
0.510
0.860
0.958
Recall/Completeness
0.984
0.935
0.442
0.860
0.698
0.463
F1
0.958
0.951
0.538
0.640
0.770
0.624

Overall accuracy 0.846

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
0.8510
0.0427
0.0826
0.0199
0.0036
0.0002
building
0.0179
0.9516
0.0263
0.0033
0.0009
0.0000
low_veg
0.0442
0.0124
0.7545
0.1885
0.0004
0.0000
tree
0.0248
0.0072
0.1528
0.8150
0.0001
0.0000
car
0.1834
0.0438
0.0117
0.0074
0.7537
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.903
0.941
0.746
0.784
0.816
--
Recall/Completeness
0.851
0.952
0.754
0.815
0.754
--
F1
0.876
0.946
0.750
0.799
0.783
--

Overall accuracy 0.845

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
0.8972
0.0367
0.0571
0.0071
0.0016
0.0002
building
0.0111
0.9755
0.0122
0.0005
0.0007
0.0000
low_veg
0.0316
0.0061
0.7799
0.1820
0.0003
0.0000
tree
0.0069
0.0025
0.1219
0.8687
0.0000
0.0000
car
0.1008
0.0508
0.0077
0.0013
0.8393
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.944
0.960
0.808
0.810
0.879
--
Recall/Completeness
0.897
0.976
0.780
0.869
0.839
--
F1
0.920
0.968
0.794
0.838
0.859
--

Overall accuracy 0.884

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
0.9020
0.0327
0.0422
0.0134
0.0064
0.0033
building
0.0188
0.9395
0.0358
0.0051
0.0008
0.0000
low_veg
0.0579
0.0208
0.8194
0.1003
0.0007
0.0008
tree
0.0283
0.0030
0.1348
0.8332
0.0003
0.0003
car
0.1908
0.0224
0.0107
0.0045
0.7642
0.0074
clutter
--
--
--
--
--
--
Precision/Correctness
0.895
0.939
0.831
0.831
0.770
--
Recall/Completeness
0.902
0.939
0.819
0.833
0.764
--
F1
0.899
0.939
0.825
0.832
0.767
--

Overall accuracy 0.874

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
0.9286
0.0304
0.0287
0.0050
0.0039
0.0035
building
0.0128
0.9654
0.0193
0.0018
0.0007
0.0000
low_veg
0.0444
0.0167
0.8483
0.0894
0.0004
0.0008
tree
0.0141
0.0008
0.1085
0.8760
0.0002
0.0003
car
0.1075
0.0219
0.0035
0.0014
0.8569
0.0088
clutter
--
--
--
--
--
--
Precision/Correctness
0.929
0.951
0.875
0.862
0.820
--
Recall/Completeness
0.929
0.965
0.848
0.876
0.857
--
F1
0.929
0.958
0.862
0.869
0.838
--

Overall accuracy 0.906

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image