Detailled semantic labeling (2D) result


Name D. Marmanis et al.
Affiliation DLR-DFD, Germany
Abbreviation DLR_10
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
0.030
0.044
0.018
0.002
0.000
building
0.036
0.936
0.022
0.006
0.000
0.000
low_veg
0.049
0.022
0.789
0.139
0.000
0.000
tree
0.016
0.005
0.083
0.895
0.000
0.000
car
0.341
0.059
0.008
0.007
0.584
0.000
clutter
0.609
0.354
0.014
0.010
0.013
0.000
Precision/Correctness
0.883
0.930
0.821
0.845
0.899
-nan
Recall/Completeness
0.905
0.936
0.789
0.895
0.584
0.000
F1
0.894
0.933
0.805
0.870
0.708
-nan

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
0.932
0.022
0.036
0.009
0.001
0.000
building
0.027
0.954
0.015
0.003
0.000
0.000
low_veg
0.037
0.014
0.825
0.123
0.000
0.000
tree
0.007
0.002
0.067
0.925
0.000
0.000
car
0.229
0.069
0.006
0.004
0.692
0.000
clutter
0.592
0.380
0.011
0.003
0.014
0.000
Precision/Correctness
0.915
0.950
0.857
0.876
0.929
-nan
Recall/Completeness
0.932
0.954
0.825
0.925
0.692
0.000
F1
0.923
0.952
0.841
0.900
0.793
-nan

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
0.9473
0.0106
0.0261
0.0130
0.0030
0.0000
building
0.0344
0.9407
0.0213
0.0035
0.0000
0.0000
low_veg
0.0748
0.0146
0.7472
0.1634
0.0000
0.0000
tree
0.0263
0.0045
0.0789
0.8902
0.0000
0.0000
car
0.3685
0.0183
0.0006
0.0017
0.6108
0.0000
clutter
0.9783
0.0003
0.0000
0.0214
0.0000
0.0000
Precision/Correctness
0.807
0.966
0.855
0.832
0.925
-nan
Recall/Completeness
0.947
0.941
0.747
0.890
0.611
0.000
F1
0.872
0.953
0.798
0.860
0.736
-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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9714
0.0068
0.0157
0.0039
0.0022
0.0000
building
0.0269
0.9605
0.0119
0.0007
0.0000
0.0000
low_veg
0.0581
0.0061
0.7802
0.1555
0.0000
0.0000
tree
0.0120
0.0014
0.0655
0.9211
0.0000
0.0000
car
0.2381
0.0209
0.0001
0.0002
0.7407
0.0000
clutter
0.9951
0.0003
0.0000
0.0046
0.0000
0.0000
Precision/Correctness
0.840
0.984
0.890
0.856
0.941
-nan
Recall/Completeness
0.971
0.960
0.780
0.921
0.741
0.000
F1
0.901
0.972
0.832
0.888
0.829
-nan

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9395
0.0456
0.0089
0.0037
0.0023
0.0000
building
0.0512
0.9371
0.0091
0.0022
0.0004
0.0000
low_veg
0.1630
0.0438
0.5049
0.2883
0.0000
0.0000
tree
0.0519
0.0159
0.0497
0.8825
0.0001
0.0000
car
0.3888
0.0042
0.0003
0.0005
0.6062
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.861
0.970
0.690
0.736
0.910
--
Recall/Completeness
0.939
0.937
0.505
0.882
0.606
--
F1
0.899
0.953
0.583
0.803
0.728
--

Overall accuracy 0.910

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
0.9659
0.0254
0.0067
0.0012
0.0008
0.0000
building
0.0368
0.9536
0.0081
0.0011
0.0004
0.0000
low_veg
0.1165
0.0124
0.5550
0.3161
0.0000
0.0000
tree
0.0206
0.0056
0.0446
0.9291
0.0000
0.0000
car
0.2689
0.0009
0.0000
0.0000
0.7302
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.901
0.986
0.718
0.762
0.947
--
Recall/Completeness
0.966
0.954
0.555
0.929
0.730
--
F1
0.932
0.970
0.626
0.837
0.825
--

Overall accuracy 0.937

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
0.8929
0.0481
0.0459
0.0110
0.0019
0.0000
building
0.0483
0.9327
0.0147
0.0041
0.0002
0.0000
low_veg
0.0732
0.0250
0.7430
0.1588
0.0000
0.0000
tree
0.0136
0.0033
0.0488
0.9342
0.0000
0.0000
car
0.4212
0.0099
0.0002
0.0015
0.5673
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.868
0.951
0.721
0.927
0.943
--
Recall/Completeness
0.893
0.933
0.743
0.934
0.567
--
F1
0.880
0.942
0.732
0.931
0.708
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9272
0.0329
0.0361
0.0032
0.0006
0.0000
building
0.0382
0.9503
0.0103
0.0010
0.0001
0.0000
low_veg
0.0595
0.0105
0.7938
0.1362
0.0000
0.0000
tree
0.0040
0.0010
0.0382
0.9568
0.0000
0.0000
car
0.2979
0.0067
0.0000
0.0000
0.6954
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.907
0.971
0.764
0.952
0.978
--
Recall/Completeness
0.927
0.950
0.794
0.957
0.695
--
F1
0.917
0.961
0.778
0.954
0.813
--

Overall accuracy 0.928

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
0.8837
0.0326
0.0711
0.0112
0.0013
0.0000
building
0.0303
0.9459
0.0191
0.0046
0.0001
0.0000
low_veg
0.0443
0.0125
0.8934
0.0497
0.0000
0.0000
tree
0.0196
0.0060
0.0680
0.9063
0.0000
0.0000
car
0.3707
0.0158
0.0133
0.0063
0.5939
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.912
0.938
0.803
0.937
0.958
--
Recall/Completeness
0.884
0.946
0.893
0.906
0.594
--
F1
0.898
0.942
0.846
0.922
0.733
--

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
0.9107
0.0263
0.0600
0.0028
0.0002
0.0000
building
0.0202
0.9676
0.0102
0.0019
0.0001
0.0000
low_veg
0.0293
0.0053
0.9275
0.0379
0.0000
0.0000
tree
0.0071
0.0025
0.0475
0.9429
0.0000
0.0000
car
0.2520
0.0177
0.0054
0.0037
0.7212
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.949
0.958
0.846
0.963
0.992
--
Recall/Completeness
0.911
0.968
0.928
0.943
0.721
--
F1
0.929
0.963
0.885
0.953
0.835
--

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
0.8577
0.0625
0.0541
0.0243
0.0014
0.0000
building
0.0219
0.9506
0.0122
0.0152
0.0000
0.0000
low_veg
0.0537
0.0298
0.7439
0.1726
0.0000
0.0000
tree
0.0115
0.0069
0.0635
0.9181
0.0000
0.0000
car
0.3516
0.0255
0.0029
0.0106
0.6094
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.943
0.881
0.783
0.814
0.890
--
Recall/Completeness
0.858
0.951
0.744
0.918
0.609
--
F1
0.898
0.915
0.763
0.863
0.724
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8912
0.0504
0.0465
0.0114
0.0006
0.0000
building
0.0174
0.9641
0.0067
0.0118
0.0000
0.0000
low_veg
0.0510
0.0179
0.7884
0.1426
0.0000
0.0000
tree
0.0038
0.0030
0.0475
0.9458
0.0000
0.0000
car
0.2616
0.0309
0.0000
0.0035
0.7039
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.958
0.910
0.821
0.858
0.943
--
Recall/Completeness
0.891
0.964
0.788
0.946
0.704
--
F1
0.924
0.936
0.805
0.900
0.806
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8783
0.0298
0.0513
0.0387
0.0018
0.0000
building
0.0553
0.9299
0.0085
0.0063
0.0000
0.0000
low_veg
0.0360
0.0273
0.7587
0.1780
0.0001
0.0000
tree
0.0074
0.0066
0.0909
0.8951
0.0000
0.0000
car
0.2327
0.0012
0.0105
0.0116
0.7439
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.834
0.924
0.838
0.850
0.930
--
Recall/Completeness
0.878
0.930
0.759
0.895
0.744
--
F1
0.855
0.927
0.796
0.872
0.827
--

Overall accuracy 0.859

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
0.9082
0.0261
0.0439
0.0217
0.0001
0.0000
building
0.0566
0.9393
0.0021
0.0020
0.0000
0.0000
low_veg
0.0338
0.0183
0.7949
0.1529
0.0000
0.0000
tree
0.0029
0.0023
0.0760
0.9188
0.0000
0.0000
car
0.1178
0.0016
0.0084
0.0045
0.8677
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.854
0.949
0.866
0.879
0.992
--
Recall/Completeness
0.908
0.939
0.795
0.919
0.868
--
F1
0.880
0.944
0.829
0.899
0.926
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8470
0.0336
0.0788
0.0358
0.0048
0.0000
building
0.0074
0.9664
0.0208
0.0054
0.0000
0.0000
low_veg
0.0502
0.0471
0.7568
0.1457
0.0001
0.0000
tree
0.0192
0.0123
0.1028
0.8657
0.0000
0.0000
car
0.1922
0.0000
0.0000
0.0071
0.8006
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.902
0.877
0.774
0.862
0.653
--
Recall/Completeness
0.847
0.966
0.757
0.866
0.801
--
F1
0.874
0.920
0.765
0.864
0.719
--

Overall accuracy 0.851

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
0.8674
0.0291
0.0741
0.0248
0.0045
0.0000
building
0.0036
0.9839
0.0111
0.0014
0.0000
0.0000
low_veg
0.0451
0.0330
0.7899
0.1320
0.0000
0.0000
tree
0.0124
0.0056
0.0907
0.8913
0.0000
0.0000
car
0.0718
0.0000
0.0000
0.0029
0.9253
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.926
0.910
0.797
0.886
0.633
--
Recall/Completeness
0.867
0.984
0.790
0.891
0.925
--
F1
0.896
0.946
0.794
0.889
0.752
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8037
0.0339
0.1032
0.0572
0.0021
0.0000
building
0.0145
0.9566
0.0159
0.0129
0.0000
0.0000
low_veg
0.0131
0.0164
0.8325
0.1378
0.0002
0.0000
tree
0.0152
0.0049
0.0942
0.8857
0.0000
0.0000
car
0.2334
0.0027
0.0022
0.0188
0.7428
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.908
0.935
0.822
0.862
0.812
--
Recall/Completeness
0.804
0.957
0.832
0.886
0.743
--
F1
0.853
0.945
0.827
0.874
0.776
--

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_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8411
0.0291
0.0933
0.0356
0.0009
0.0000
building
0.0122
0.9766
0.0062
0.0050
0.0000
0.0000
low_veg
0.0103
0.0100
0.8582
0.1213
0.0002
0.0000
tree
0.0092
0.0014
0.0809
0.9085
0.0000
0.0000
car
0.1407
0.0001
0.0000
0.0059
0.8533
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.935
0.957
0.850
0.888
0.877
--
Recall/Completeness
0.841
0.977
0.858
0.909
0.853
--
F1
0.886
0.967
0.854
0.898
0.865
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8556
0.0305
0.0586
0.0527
0.0027
0.0000
building
0.0201
0.9436
0.0300
0.0063
0.0000
0.0000
low_veg
0.0285
0.0261
0.7588
0.1866
0.0000
0.0000
tree
0.0055
0.0015
0.0405
0.9524
0.0000
0.0000
car
0.2223
0.0534
0.0321
0.0189
0.6733
0.0000
clutter
0.0094
0.9875
0.0017
0.0014
0.0000
0.0000
Precision/Correctness
0.934
0.799
0.814
0.891
0.839
-nan
Recall/Completeness
0.856
0.944
0.759
0.952
0.673
0.000
F1
0.893
0.865
0.786
0.921
0.747
-nan

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8919
0.0200
0.0550
0.0314
0.0017
0.0000
building
0.0190
0.9547
0.0247
0.0016
0.0000
0.0000
low_veg
0.0219
0.0161
0.8081
0.1539
0.0000
0.0000
tree
0.0019
0.0003
0.0299
0.9678
0.0000
0.0000
car
0.1189
0.0646
0.0254
0.0092
0.7818
0.0000
clutter
0.0073
0.9922
0.0002
0.0003
0.0000
0.0000
Precision/Correctness
0.954
0.818
0.843
0.920
0.870
-nan
Recall/Completeness
0.892
0.955
0.808
0.968
0.782
0.000
F1
0.922
0.881
0.825
0.943
0.824
-nan

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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8904
0.0295
0.0399
0.0364
0.0038
0.0000
building
0.0149
0.9682
0.0093
0.0076
0.0001
0.0000
low_veg
0.0721
0.0341
0.7409
0.1528
0.0000
0.0000
tree
0.0167
0.0040
0.0498
0.9295
0.0000
0.0000
car
0.3136
0.0224
0.0061
0.0120
0.6458
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.918
0.907
0.838
0.859
0.849
--
Recall/Completeness
0.890
0.968
0.741
0.929
0.646
--
F1
0.904
0.937
0.787
0.893
0.734
--

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
0.9229
0.0218
0.0343
0.0188
0.0021
0.0000
building
0.0098
0.9836
0.0030
0.0035
0.0000
0.0000
low_veg
0.0682
0.0209
0.7894
0.1215
0.0000
0.0000
tree
0.0055
0.0011
0.0345
0.9589
0.0000
0.0000
car
0.2086
0.0247
0.0040
0.0036
0.7590
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.942
0.939
0.877
0.900
0.898
--
Recall/Completeness
0.923
0.984
0.789
0.959
0.759
--
F1
0.932
0.961
0.831
0.929
0.823
--

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_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8123
0.0412
0.1001
0.0424
0.0040
0.0000
building
0.0559
0.8893
0.0422
0.0126
0.0000
0.0000
low_veg
0.0169
0.0320
0.8706
0.0804
0.0001
0.0000
tree
0.0095
0.0058
0.1436
0.8410
0.0000
0.0000
car
0.3038
0.0000
0.0026
0.0087
0.6848
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.849
0.890
0.859
0.837
0.809
--
Recall/Completeness
0.812
0.889
0.871
0.841
0.685
--
F1
0.830
0.890
0.865
0.839
0.742
--

Overall accuracy 0.857

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
0.8467
0.0299
0.0906
0.0299
0.0028
0.0000
building
0.0553
0.9018
0.0335
0.0094
0.0000
0.0000
low_veg
0.0139
0.0288
0.8880
0.0692
0.0001
0.0000
tree
0.0047
0.0024
0.1210
0.8719
0.0000
0.0000
car
0.1890
0.0000
0.0000
0.0045
0.8065
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.870
0.911
0.882
0.861
0.831
--
Recall/Completeness
0.847
0.902
0.888
0.872
0.806
--
F1
0.858
0.906
0.885
0.867
0.818
--

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
0.9446
0.0217
0.0226
0.0075
0.0036
0.0000
building
0.0662
0.9122
0.0152
0.0060
0.0003
0.0000
low_veg
0.0918
0.0218
0.6792
0.2072
0.0000
0.0000
tree
0.0107
0.0042
0.0329
0.9522
0.0000
0.0000
car
0.3908
0.0099
0.0035
0.0079
0.5879
0.0000
clutter
0.8133
0.0208
0.1116
0.0043
0.0501
0.0000
Precision/Correctness
0.841
0.964
0.891
0.678
0.932
-nan
Recall/Completeness
0.945
0.912
0.679
0.952
0.588
0.000
F1
0.890
0.937
0.771
0.792
0.721
-nan

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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9673
0.0128
0.0162
0.0029
0.0008
0.0000
building
0.0542
0.9327
0.0099
0.0029
0.0003
0.0000
low_veg
0.0730
0.0089
0.7234
0.1947
0.0000
0.0000
tree
0.0042
0.0012
0.0200
0.9746
0.0000
0.0000
car
0.2781
0.0094
0.0019
0.0056
0.7048
0.0000
clutter
0.8295
0.0128
0.1004
0.0014
0.0559
0.0000
Precision/Correctness
0.879
0.982
0.926
0.710
0.974
-nan
Recall/Completeness
0.967
0.933
0.723
0.975
0.705
0.000
F1
0.921
0.957
0.812
0.821
0.818
-nan

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9416
0.0188
0.0227
0.0139
0.0029
0.0000
building
0.0237
0.9591
0.0101
0.0071
0.0000
0.0000
low_veg
0.0704
0.0224
0.6102
0.2969
0.0001
0.0000
tree
0.0030
0.0017
0.0199
0.9753
0.0000
0.0000
car
0.3361
0.0035
0.0044
0.0104
0.6456
0.0000
clutter
0.5705
0.3573
0.0290
0.0189
0.0243
0.0000
Precision/Correctness
0.912
0.922
0.854
0.671
0.814
-nan
Recall/Completeness
0.942
0.959
0.610
0.975
0.646
0.000
F1
0.926
0.940
0.712
0.795
0.720
-nan

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9619
0.0128
0.0167
0.0064
0.0022
0.0000
building
0.0173
0.9734
0.0055
0.0037
0.0000
0.0000
low_veg
0.0483
0.0091
0.6589
0.2837
0.0000
0.0000
tree
0.0005
0.0003
0.0145
0.9846
0.0000
0.0000
car
0.2284
0.0019
0.0052
0.0072
0.7573
0.0000
clutter
0.5515
0.3859
0.0271
0.0086
0.0270
0.0000
Precision/Correctness
0.935
0.941
0.890
0.709
0.823
-nan
Recall/Completeness
0.962
0.973
0.659
0.985
0.757
0.000
F1
0.948
0.957
0.757
0.825
0.789
-nan

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9597
0.0197
0.0142
0.0044
0.0021
0.0000
building
0.0281
0.9574
0.0107
0.0030
0.0007
0.0000
low_veg
0.0387
0.0054
0.9361
0.0198
0.0000
0.0000
tree
0.0346
0.0032
0.0885
0.8736
0.0001
0.0000
car
0.2261
0.3047
0.0064
0.0100
0.4528
0.0000
clutter
0.5369
0.4168
0.0276
0.0099
0.0089
0.0000
Precision/Correctness
0.895
0.887
0.876
0.942
0.924
-nan
Recall/Completeness
0.960
0.957
0.936
0.874
0.453
0.000
F1
0.926
0.921
0.905
0.906
0.608
-nan

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9735
0.0141
0.0108
0.0008
0.0009
0.0000
building
0.0164
0.9773
0.0051
0.0005
0.0006
0.0000
low_veg
0.0214
0.0032
0.9621
0.0132
0.0000
0.0000
tree
0.0110
0.0006
0.0703
0.9180
0.0000
0.0000
car
0.1308
0.3430
0.0016
0.0049
0.5198
0.0000
clutter
0.4915
0.4730
0.0220
0.0027
0.0108
0.0000
Precision/Correctness
0.932
0.904
0.911
0.974
0.947
-nan
Recall/Completeness
0.973
0.977
0.962
0.918
0.520
0.000
F1
0.952
0.939
0.936
0.945
0.671
-nan

Overall accuracy 0.924

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
0.9740
0.0135
0.0092
0.0026
0.0006
0.0000
building
0.0297
0.8742
0.0819
0.0130
0.0012
0.0000
low_veg
0.0853
0.0103
0.8347
0.0696
0.0000
0.0000
tree
0.0392
0.0031
0.1594
0.7984
0.0000
0.0000
car
0.4268
0.0191
0.0111
0.0056
0.5374
0.0000
clutter
0.7863
0.1895
0.0026
0.0015
0.0201
0.0000
Precision/Correctness
0.836
0.935
0.744
0.843
0.804
-nan
Recall/Completeness
0.974
0.874
0.835
0.798
0.537
0.000
F1
0.900
0.904
0.787
0.820
0.644
-nan

Overall accuracy 0.847

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
0.9854
0.0093
0.0044
0.0004
0.0004
0.0000
building
0.0202
0.8878
0.0799
0.0109
0.0012
0.0000
low_veg
0.0554
0.0045
0.8684
0.0717
0.0000
0.0000
tree
0.0158
0.0004
0.1436
0.8402
0.0000
0.0000
car
0.3190
0.0220
0.0062
0.0064
0.6464
0.0000
clutter
0.7731
0.2024
0.0008
0.0013
0.0224
0.0000
Precision/Correctness
0.868
0.948
0.764
0.857
0.798
-nan
Recall/Completeness
0.985
0.888
0.868
0.840
0.646
0.000
F1
0.923
0.917
0.813
0.848
0.714
-nan

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8572
0.0325
0.0930
0.0165
0.0007
0.0000
building
0.0307
0.9439
0.0195
0.0055
0.0003
0.0000
low_veg
0.0312
0.0108
0.7589
0.1992
0.0000
0.0000
tree
0.0235
0.0048
0.1238
0.8478
0.0000
0.0000
car
0.3903
0.0020
0.0002
0.0078
0.5997
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.898
0.954
0.766
0.783
0.943
--
Recall/Completeness
0.857
0.944
0.759
0.848
0.600
--
F1
0.877
0.949
0.762
0.814
0.733
--

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_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9046
0.0272
0.0638
0.0042
0.0002
0.0000
building
0.0209
0.9695
0.0078
0.0014
0.0003
0.0000
low_veg
0.0195
0.0035
0.7829
0.1940
0.0000
0.0000
tree
0.0065
0.0008
0.0925
0.9002
0.0000
0.0000
car
0.2735
0.0014
0.0000
0.0013
0.7238
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.942
0.972
0.831
0.808
0.964
--
Recall/Completeness
0.905
0.969
0.783
0.900
0.724
--
F1
0.923
0.971
0.806
0.851
0.827
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9015
0.0268
0.0558
0.0143
0.0016
0.0000
building
0.0211
0.9276
0.0446
0.0065
0.0003
0.0000
low_veg
0.0443
0.0160
0.8629
0.0768
0.0000
0.0000
tree
0.0182
0.0031
0.1189
0.8598
0.0000
0.0000
car
0.3375
0.0202
0.0468
0.0073
0.5882
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.908
0.950
0.829
0.861
0.919
--
Recall/Completeness
0.902
0.928
0.863
0.860
0.588
--
F1
0.905
0.938
0.846
0.861
0.717
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9298
0.0240
0.0412
0.0047
0.0004
0.0000
building
0.0133
0.9536
0.0297
0.0031
0.0003
0.0000
low_veg
0.0297
0.0115
0.8966
0.0622
0.0000
0.0000
tree
0.0062
0.0005
0.0930
0.9003
0.0000
0.0000
car
0.2292
0.0228
0.0433
0.0017
0.7030
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.945
0.962
0.871
0.899
0.965
--
Recall/Completeness
0.930
0.954
0.897
0.900
0.703
--
F1
0.937
0.958
0.883
0.899
0.814
--

Overall accuracy 0.920

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image