Detailled semantic labeling (2D) result


Name D. Marmanis et al. (unaries CNN)
Affiliation DLR-DFD, Germany
Abbreviation DLR_3
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
0.882
0.046
0.047
0.023
0.003
0.000
building
0.037
0.930
0.025
0.008
0.000
0.000
low_veg
0.045
0.038
0.737
0.180
0.000
0.000
tree
0.014
0.006
0.077
0.903
0.000
0.000
car
0.275
0.071
0.008
0.010
0.636
0.000
clutter
0.564
0.392
0.016
0.010
0.017
0.000
Precision/Correctness
0.887
0.901
0.811
0.812
0.874
-nan
Recall/Completeness
0.882
0.930
0.737
0.903
0.636
0.000
F1
0.884
0.916
0.772
0.855
0.736
-nan

Overall accuracy 0.857

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
0.912
0.038
0.038
0.011
0.002
0.000
building
0.030
0.948
0.018
0.004
0.000
0.000
low_veg
0.035
0.031
0.769
0.165
0.000
0.000
tree
0.006
0.002
0.062
0.930
0.000
0.000
car
0.167
0.081
0.006
0.006
0.740
0.000
clutter
0.550
0.415
0.013
0.004
0.019
0.000
Precision/Correctness
0.917
0.921
0.847
0.843
0.906
-nan
Recall/Completeness
0.912
0.948
0.769
0.930
0.740
0.000
F1
0.914
0.934
0.806
0.884
0.815
-nan

Overall accuracy 0.886
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9354
0.0147
0.0290
0.0166
0.0042
0.0000
building
0.0320
0.9413
0.0230
0.0037
0.0000
0.0000
low_veg
0.0758
0.0162
0.7188
0.1891
0.0000
0.0000
tree
0.0249
0.0055
0.0726
0.8970
0.0001
0.0000
car
0.2980
0.0179
0.0005
0.0037
0.6799
0.0000
clutter
0.9734
0.0039
0.0000
0.0227
0.0000
0.0000
Precision/Correctness
0.810
0.958
0.852
0.811
0.907
-nan
Recall/Completeness
0.935
0.941
0.719
0.897
0.680
0.000
F1
0.868
0.950
0.780
0.852
0.777
-nan

Overall accuracy 0.853

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
0.9633
0.0105
0.0181
0.0050
0.0031
0.0000
building
0.0241
0.9615
0.0138
0.0006
0.0000
0.0000
low_veg
0.0611
0.0080
0.7511
0.1798
0.0000
0.0000
tree
0.0117
0.0024
0.0603
0.9255
0.0001
0.0000
car
0.1660
0.0194
0.0000
0.0020
0.8126
0.0000
clutter
0.9913
0.0035
0.0000
0.0052
0.0000
0.0000
Precision/Correctness
0.841
0.977
0.888
0.838
0.926
-nan
Recall/Completeness
0.963
0.961
0.751
0.925
0.813
0.000
F1
0.898
0.969
0.814
0.880
0.866
-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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9003
0.0797
0.0110
0.0054
0.0036
0.0000
building
0.0414
0.9469
0.0089
0.0024
0.0003
0.0000
low_veg
0.1771
0.0410
0.4651
0.3163
0.0005
0.0000
tree
0.0457
0.0138
0.0489
0.8911
0.0005
0.0000
car
0.3388
0.0095
0.0005
0.0007
0.6505
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.873
0.954
0.662
0.715
0.881
--
Recall/Completeness
0.900
0.947
0.465
0.891
0.651
--
F1
0.886
0.950
0.546
0.793
0.748
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9311
0.0566
0.0085
0.0019
0.0019
0.0000
building
0.0293
0.9613
0.0078
0.0012
0.0003
0.0000
low_veg
0.1357
0.0113
0.5112
0.3415
0.0002
0.0000
tree
0.0182
0.0052
0.0428
0.9338
0.0001
0.0000
car
0.2237
0.0050
0.0000
0.0000
0.7712
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.911
0.971
0.694
0.746
0.918
--
Recall/Completeness
0.931
0.961
0.511
0.934
0.771
--
F1
0.921
0.966
0.589
0.829
0.838
--

Overall accuracy 0.930

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
0.8790
0.0595
0.0440
0.0150
0.0025
0.0000
building
0.0437
0.9368
0.0141
0.0054
0.0000
0.0000
low_veg
0.0620
0.0304
0.7250
0.1826
0.0001
0.0000
tree
0.0120
0.0031
0.0428
0.9421
0.0000
0.0000
car
0.3528
0.0602
0.0018
0.0050
0.5802
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.881
0.939
0.732
0.915
0.935
--
Recall/Completeness
0.879
0.937
0.725
0.942
0.580
--
F1
0.880
0.938
0.729
0.928
0.716
--

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
0.9171
0.0435
0.0339
0.0046
0.0009
0.0000
building
0.0339
0.9541
0.0098
0.0021
0.0000
0.0000
low_veg
0.0481
0.0183
0.7759
0.1577
0.0000
0.0000
tree
0.0038
0.0009
0.0325
0.9628
0.0000
0.0000
car
0.2342
0.0609
0.0003
0.0020
0.7026
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.918
0.960
0.778
0.943
0.975
--
Recall/Completeness
0.917
0.954
0.776
0.963
0.703
--
F1
0.918
0.957
0.777
0.953
0.817
--

Overall accuracy 0.927

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
0.8694
0.0388
0.0757
0.0140
0.0021
0.0000
building
0.0233
0.9500
0.0212
0.0053
0.0003
0.0000
low_veg
0.0446
0.0131
0.8772
0.0648
0.0003
0.0000
tree
0.0167
0.0051
0.0638
0.9144
0.0000
0.0000
car
0.3280
0.0257
0.0150
0.0070
0.6243
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.920
0.930
0.796
0.922
0.930
--
Recall/Completeness
0.869
0.950
0.877
0.914
0.624
--
F1
0.894
0.940
0.835
0.918
0.747
--

Overall accuracy 0.896

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
0.8992
0.0325
0.0639
0.0037
0.0007
0.0000
building
0.0142
0.9716
0.0118
0.0022
0.0002
0.0000
low_veg
0.0317
0.0060
0.9117
0.0503
0.0002
0.0000
tree
0.0059
0.0022
0.0453
0.9465
0.0000
0.0000
car
0.2151
0.0270
0.0097
0.0047
0.7435
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.953
0.950
0.840
0.952
0.970
--
Recall/Completeness
0.899
0.972
0.912
0.947
0.744
--
F1
0.926
0.961
0.874
0.949
0.842
--

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_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8254
0.0877
0.0580
0.0273
0.0016
0.0000
building
0.0164
0.9539
0.0168
0.0129
0.0000
0.0000
low_veg
0.0451
0.0367
0.7446
0.1736
0.0000
0.0000
tree
0.0092
0.0066
0.0641
0.9201
0.0000
0.0000
car
0.2761
0.0328
0.0012
0.0088
0.6811
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.952
0.846
0.771
0.812
0.889
--
Recall/Completeness
0.825
0.954
0.745
0.920
0.681
--
F1
0.884
0.897
0.757
0.862
0.771
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8617
0.0751
0.0498
0.0128
0.0006
0.0000
building
0.0125
0.9671
0.0112
0.0092
0.0000
0.0000
low_veg
0.0416
0.0260
0.7905
0.1420
0.0000
0.0000
tree
0.0037
0.0027
0.0479
0.9457
0.0000
0.0000
car
0.1826
0.0336
0.0000
0.0025
0.7812
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.966
0.873
0.810
0.859
0.946
--
Recall/Completeness
0.862
0.967
0.791
0.946
0.781
--
F1
0.911
0.918
0.800
0.900
0.856
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8647
0.0323
0.0564
0.0440
0.0027
0.0000
building
0.1457
0.8362
0.0112
0.0068
0.0000
0.0000
low_veg
0.0311
0.0289
0.7529
0.1870
0.0002
0.0000
tree
0.0064
0.0059
0.0881
0.8995
0.0000
0.0000
car
0.1805
0.0122
0.0259
0.0186
0.7628
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.754
0.913
0.836
0.843
0.896
--
Recall/Completeness
0.865
0.836
0.753
0.900
0.763
--
F1
0.805
0.873
0.792
0.870
0.824
--

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_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8981
0.0287
0.0480
0.0248
0.0005
0.0000
building
0.1541
0.8405
0.0033
0.0021
0.0000
0.0000
low_veg
0.0291
0.0204
0.7888
0.1616
0.0002
0.0000
tree
0.0024
0.0021
0.0732
0.9224
0.0000
0.0000
car
0.0858
0.0156
0.0244
0.0091
0.8652
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.765
0.939
0.865
0.873
0.965
--
Recall/Completeness
0.898
0.840
0.789
0.922
0.865
--
F1
0.826
0.887
0.825
0.897
0.912
--

Overall accuracy 0.865

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
0.8373
0.0437
0.0691
0.0443
0.0055
0.0000
building
0.0071
0.9610
0.0257
0.0062
0.0000
0.0000
low_veg
0.0459
0.0503
0.7288
0.1749
0.0001
0.0000
tree
0.0191
0.0116
0.1018
0.8675
0.0000
0.0000
car
0.1566
0.0028
0.0000
0.0095
0.8311
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.907
0.863
0.773
0.839
0.631
--
Recall/Completeness
0.837
0.961
0.729
0.867
0.831
--
F1
0.871
0.910
0.750
0.853
0.717
--

Overall accuracy 0.842

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
0.8601
0.0392
0.0640
0.0316
0.0051
0.0000
building
0.0034
0.9808
0.0140
0.0018
0.0000
0.0000
low_veg
0.0409
0.0372
0.7607
0.1611
0.0000
0.0000
tree
0.0127
0.0056
0.0905
0.8911
0.0000
0.0000
car
0.0488
0.0029
0.0000
0.0058
0.9425
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.930
0.894
0.798
0.863
0.609
--
Recall/Completeness
0.860
0.981
0.761
0.891
0.942
--
F1
0.894
0.935
0.779
0.877
0.740
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7705
0.0515
0.1075
0.0683
0.0021
0.0000
building
0.0093
0.9539
0.0226
0.0141
0.0001
0.0000
low_veg
0.0105
0.0298
0.7478
0.2116
0.0002
0.0000
tree
0.0131
0.0051
0.0882
0.8937
0.0000
0.0000
car
0.1676
0.0046
0.0007
0.0226
0.8045
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.923
0.901
0.807
0.812
0.815
--
Recall/Completeness
0.770
0.954
0.748
0.894
0.805
--
F1
0.840
0.927
0.776
0.851
0.810
--

Overall accuracy 0.842

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
0.8098
0.0466
0.0988
0.0438
0.0009
0.0000
building
0.0071
0.9753
0.0119
0.0057
0.0001
0.0000
low_veg
0.0079
0.0243
0.7693
0.1984
0.0002
0.0000
tree
0.0079
0.0017
0.0758
0.9146
0.0000
0.0000
car
0.0715
0.0014
0.0000
0.0132
0.9139
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.949
0.922
0.836
0.835
0.878
--
Recall/Completeness
0.810
0.975
0.769
0.915
0.914
--
F1
0.874
0.948
0.801
0.873
0.896
--

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8067
0.0602
0.0637
0.0657
0.0037
0.0000
building
0.0111
0.9509
0.0303
0.0076
0.0000
0.0000
low_veg
0.0266
0.0252
0.7415
0.2067
0.0000
0.0000
tree
0.0043
0.0015
0.0401
0.9541
0.0000
0.0000
car
0.1320
0.1561
0.0389
0.0220
0.6510
0.0000
clutter
0.0047
0.9913
0.0034
0.0006
0.0000
0.0000
Precision/Correctness
0.945
0.775
0.806
0.878
0.782
-nan
Recall/Completeness
0.807
0.951
0.742
0.954
0.651
0.000
F1
0.870
0.854
0.773
0.914
0.710
-nan

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8452
0.0499
0.0590
0.0435
0.0024
0.0000
building
0.0097
0.9629
0.0251
0.0022
0.0000
0.0000
low_veg
0.0222
0.0149
0.7898
0.1730
0.0000
0.0000
tree
0.0014
0.0003
0.0301
0.9681
0.0000
0.0000
car
0.0460
0.1885
0.0342
0.0110
0.7202
0.0000
clutter
0.0041
0.9948
0.0011
0.0000
0.0000
0.0000
Precision/Correctness
0.963
0.794
0.835
0.908
0.813
-nan
Recall/Completeness
0.845
0.963
0.790
0.968
0.720
0.000
F1
0.900
0.870
0.812
0.937
0.764
-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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8758
0.0332
0.0424
0.0443
0.0044
0.0000
building
0.0162
0.9640
0.0108
0.0089
0.0001
0.0000
low_veg
0.0684
0.0333
0.7283
0.1699
0.0000
0.0000
tree
0.0138
0.0035
0.0435
0.9393
0.0000
0.0000
car
0.2655
0.0362
0.0126
0.0188
0.6669
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.922
0.902
0.839
0.844
0.834
--
Recall/Completeness
0.876
0.964
0.728
0.939
0.667
--
F1
0.898
0.932
0.780
0.889
0.741
--

Overall accuracy 0.878

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
0.9113
0.0250
0.0371
0.0240
0.0027
0.0000
building
0.0117
0.9797
0.0044
0.0041
0.0000
0.0000
low_veg
0.0647
0.0200
0.7781
0.1372
0.0000
0.0000
tree
0.0041
0.0010
0.0299
0.9651
0.0000
0.0000
car
0.1582
0.0440
0.0113
0.0098
0.7768
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.945
0.934
0.876
0.887
0.878
--
Recall/Completeness
0.911
0.980
0.778
0.965
0.777
--
F1
0.928
0.956
0.824
0.924
0.824
--

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_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7616
0.0717
0.1076
0.0545
0.0046
0.0000
building
0.0480
0.8831
0.0585
0.0103
0.0000
0.0000
low_veg
0.0129
0.1027
0.7665
0.1177
0.0002
0.0000
tree
0.0065
0.0061
0.1196
0.8677
0.0000
0.0000
car
0.2517
0.0001
0.0026
0.0167
0.7289
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.867
0.754
0.847
0.791
0.787
--
Recall/Completeness
0.762
0.883
0.766
0.868
0.729
--
F1
0.811
0.813
0.805
0.827
0.757
--

Overall accuracy 0.813

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
0.7977
0.0611
0.0980
0.0397
0.0035
0.0000
building
0.0480
0.8960
0.0486
0.0074
0.0000
0.0000
low_veg
0.0107
0.1037
0.7803
0.1052
0.0002
0.0000
tree
0.0028
0.0026
0.0998
0.8948
0.0000
0.0000
car
0.1302
0.0000
0.0000
0.0108
0.8590
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.886
0.765
0.871
0.814
0.797
--
Recall/Completeness
0.798
0.896
0.780
0.895
0.859
--
F1
0.840
0.825
0.823
0.852
0.827
--

Overall accuracy 0.833

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
0.9317
0.0296
0.0236
0.0092
0.0059
0.0000
building
0.0583
0.9170
0.0166
0.0076
0.0005
0.0000
low_veg
0.0979
0.0255
0.6491
0.2270
0.0005
0.0000
tree
0.0100
0.0044
0.0274
0.9580
0.0002
0.0000
car
0.2954
0.0186
0.0030
0.0086
0.6745
0.0000
clutter
0.7768
0.0322
0.1210
0.0060
0.0640
0.0000
Precision/Correctness
0.851
0.954
0.885
0.656
0.904
-nan
Recall/Completeness
0.932
0.917
0.649
0.958
0.674
0.000
F1
0.890
0.935
0.749
0.779
0.772
-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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9585
0.0196
0.0168
0.0036
0.0015
0.0000
building
0.0465
0.9371
0.0117
0.0043
0.0004
0.0000
low_veg
0.0808
0.0139
0.6914
0.2133
0.0005
0.0000
tree
0.0039
0.0012
0.0158
0.9790
0.0000
0.0000
car
0.1875
0.0200
0.0018
0.0066
0.7841
0.0000
clutter
0.7975
0.0223
0.1062
0.0034
0.0707
0.0000
Precision/Correctness
0.887
0.973
0.920
0.689
0.957
-nan
Recall/Completeness
0.958
0.937
0.691
0.979
0.784
0.000
F1
0.921
0.955
0.789
0.809
0.862
-nan

Overall accuracy 0.896

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
0.9260
0.0289
0.0245
0.0161
0.0045
0.0000
building
0.0161
0.9654
0.0111
0.0074
0.0001
0.0000
low_veg
0.0578
0.0247
0.5905
0.3266
0.0004
0.0000
tree
0.0023
0.0017
0.0171
0.9790
0.0000
0.0000
car
0.2944
0.0112
0.0042
0.0120
0.6781
0.0000
clutter
0.5584
0.3694
0.0271
0.0211
0.0240
0.0000
Precision/Correctness
0.921
0.906
0.844
0.649
0.772
-nan
Recall/Completeness
0.926
0.965
0.591
0.979
0.678
0.000
F1
0.924
0.935
0.695
0.781
0.722
-nan

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9487
0.0224
0.0181
0.0074
0.0034
0.0000
building
0.0109
0.9791
0.0062
0.0038
0.0000
0.0000
low_veg
0.0366
0.0122
0.6378
0.3131
0.0003
0.0000
tree
0.0004
0.0003
0.0117
0.9876
0.0000
0.0000
car
0.1939
0.0093
0.0050
0.0084
0.7834
0.0000
clutter
0.5446
0.3930
0.0250
0.0111
0.0262
0.0000
Precision/Correctness
0.943
0.926
0.884
0.689
0.781
-nan
Recall/Completeness
0.949
0.979
0.638
0.988
0.783
0.000
F1
0.946
0.952
0.741
0.812
0.782
-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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9269
0.0520
0.0127
0.0052
0.0031
0.0000
building
0.0303
0.9535
0.0111
0.0036
0.0014
0.0000
low_veg
0.0423
0.0068
0.9234
0.0275
0.0000
0.0000
tree
0.0300
0.0029
0.0832
0.8836
0.0003
0.0000
car
0.1812
0.2849
0.0061
0.0103
0.5175
0.0000
clutter
0.4881
0.4467
0.0331
0.0101
0.0221
0.0000
Precision/Correctness
0.898
0.849
0.880
0.928
0.886
-nan
Recall/Completeness
0.927
0.954
0.923
0.884
0.518
0.000
F1
0.912
0.898
0.901
0.905
0.653
-nan

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9414
0.0474
0.0090
0.0009
0.0013
0.0000
building
0.0197
0.9723
0.0056
0.0012
0.0013
0.0000
low_veg
0.0242
0.0039
0.9524
0.0196
0.0000
0.0000
tree
0.0107
0.0006
0.0661
0.9226
0.0000
0.0000
car
0.0998
0.3159
0.0021
0.0050
0.5773
0.0000
clutter
0.4418
0.5015
0.0281
0.0032
0.0254
0.0000
Precision/Correctness
0.932
0.867
0.915
0.963
0.911
-nan
Recall/Completeness
0.941
0.972
0.952
0.923
0.577
0.000
F1
0.937
0.917
0.933
0.942
0.707
-nan

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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9260
0.0567
0.0131
0.0035
0.0008
0.0000
building
0.0495
0.8542
0.0754
0.0189
0.0021
0.0000
low_veg
0.0678
0.0265
0.4663
0.4393
0.0000
0.0000
tree
0.0321
0.0304
0.1208
0.8167
0.0000
0.0000
car
0.2934
0.1002
0.0160
0.0018
0.5885
0.0000
clutter
0.6793
0.2911
0.0038
0.0003
0.0255
0.0000
Precision/Correctness
0.839
0.843
0.643
0.535
0.767
-nan
Recall/Completeness
0.926
0.854
0.466
0.817
0.589
0.000
F1
0.880
0.849
0.540
0.646
0.666
-nan

Overall accuracy 0.769

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
0.9413
0.0501
0.0074
0.0006
0.0005
0.0000
building
0.0433
0.8642
0.0736
0.0170
0.0020
0.0000
low_veg
0.0399
0.0191
0.4656
0.4755
0.0000
0.0000
tree
0.0114
0.0284
0.1068
0.8533
0.0000
0.0000
car
0.1671
0.1156
0.0122
0.0008
0.7043
0.0000
clutter
0.6662
0.3028
0.0025
0.0000
0.0284
0.0000
Precision/Correctness
0.868
0.863
0.661
0.531
0.759
-nan
Recall/Completeness
0.941
0.864
0.466
0.853
0.704
0.000
F1
0.903
0.863
0.546
0.654
0.731
-nan

Overall accuracy 0.788

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
0.8388
0.0417
0.0953
0.0235
0.0007
0.0000
building
0.0374
0.9348
0.0205
0.0067
0.0005
0.0000
low_veg
0.0337
0.0097
0.7226
0.2340
0.0000
0.0000
tree
0.0208
0.0044
0.1102
0.8646
0.0000
0.0000
car
0.3395
0.0059
0.0001
0.0117
0.6428
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.891
0.947
0.765
0.755
0.935
--
Recall/Completeness
0.839
0.935
0.723
0.865
0.643
--
F1
0.864
0.941
0.743
0.806
0.762
--

Overall accuracy 0.840

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
0.8911
0.0361
0.0666
0.0060
0.0001
0.0000
building
0.0275
0.9617
0.0084
0.0018
0.0005
0.0000
low_veg
0.0222
0.0032
0.7490
0.2257
0.0000
0.0000
tree
0.0065
0.0009
0.0828
0.9098
0.0000
0.0000
car
0.2109
0.0061
0.0000
0.0045
0.7785
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.932
0.965
0.830
0.784
0.962
--
Recall/Completeness
0.891
0.962
0.749
0.910
0.778
--
F1
0.911
0.963
0.787
0.842
0.861
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8872
0.0289
0.0632
0.0186
0.0020
0.0000
building
0.0225
0.9104
0.0570
0.0098
0.0004
0.0000
low_veg
0.0346
0.0554
0.7871
0.1228
0.0000
0.0000
tree
0.0140
0.0032
0.1116
0.8711
0.0000
0.0000
car
0.2653
0.0460
0.0345
0.0188
0.6353
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.920
0.905
0.805
0.802
0.903
--
Recall/Completeness
0.887
0.910
0.787
0.871
0.635
--
F1
0.903
0.907
0.796
0.835
0.746
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9198
0.0259
0.0475
0.0060
0.0007
0.0000
building
0.0153
0.9362
0.0426
0.0056
0.0003
0.0000
low_veg
0.0203
0.0559
0.8152
0.1085
0.0000
0.0000
tree
0.0046
0.0009
0.0894
0.9052
0.0000
0.0000
car
0.1627
0.0540
0.0266
0.0131
0.7436
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.955
0.915
0.846
0.838
0.950
--
Recall/Completeness
0.920
0.936
0.815
0.905
0.744
--
F1
0.937
0.925
0.830
0.870
0.834
--

Overall accuracy 0.891

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image