Detailled semantic labeling (2D) result


Name D. Marmanis et al.
Affiliation DLR-DFD, Germany
Abbreviation DLR_7
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
0.894
0.036
0.046
0.021
0.003
0.000
building
0.051
0.918
0.024
0.007
0.000
0.000
low_veg
0.046
0.024
0.770
0.160
0.000
0.000
tree
0.014
0.005
0.078
0.903
0.000
0.000
car
0.281
0.065
0.008
0.008
0.638
0.000
clutter
0.586
0.373
0.015
0.010
0.016
0.000
Precision/Correctness
0.876
0.921
0.820
0.828
0.885
-nan
Recall/Completeness
0.894
0.918
0.770
0.903
0.638
0.000
F1
0.885
0.920
0.794
0.864
0.741
-nan

Overall accuracy 0.864

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
0.924
0.027
0.037
0.011
0.002
0.000
building
0.045
0.935
0.017
0.004
0.000
0.000
low_veg
0.035
0.016
0.806
0.143
0.000
0.000
tree
0.006
0.002
0.063
0.930
0.000
0.000
car
0.172
0.074
0.006
0.005
0.744
0.000
clutter
0.570
0.398
0.012
0.003
0.017
0.000
Precision/Correctness
0.903
0.942
0.855
0.860
0.917
-nan
Recall/Completeness
0.924
0.935
0.806
0.930
0.744
0.000
F1
0.914
0.938
0.830
0.893
0.821
-nan

Overall accuracy 0.894
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9397
0.0126
0.0281
0.0156
0.0040
0.0000
building
0.0325
0.9420
0.0220
0.0035
0.0000
0.0000
low_veg
0.0754
0.0142
0.7268
0.1836
0.0000
0.0000
tree
0.0246
0.0045
0.0750
0.8959
0.0001
0.0000
car
0.2994
0.0173
0.0004
0.0029
0.6800
0.0000
clutter
0.9752
0.0029
0.0000
0.0218
0.0000
0.0000
Precision/Correctness
0.811
0.964
0.853
0.815
0.912
-nan
Recall/Completeness
0.940
0.942
0.727
0.896
0.680
0.000
F1
0.871
0.953
0.785
0.854
0.779
-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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9664
0.0085
0.0176
0.0045
0.0030
0.0000
building
0.0249
0.9618
0.0128
0.0005
0.0000
0.0000
low_veg
0.0599
0.0061
0.7595
0.1746
0.0000
0.0000
tree
0.0114
0.0014
0.0626
0.9246
0.0000
0.0000
car
0.1636
0.0186
0.0000
0.0014
0.8164
0.0000
clutter
0.9927
0.0025
0.0000
0.0047
0.0000
0.0000
Precision/Correctness
0.842
0.982
0.888
0.842
0.930
-nan
Recall/Completeness
0.966
0.962
0.759
0.925
0.816
0.000
F1
0.900
0.972
0.819
0.882
0.870
-nan

Overall accuracy 0.885

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
0.9168
0.0646
0.0107
0.0049
0.0030
0.0000
building
0.0428
0.9456
0.0088
0.0024
0.0005
0.0000
low_veg
0.1711
0.0393
0.4767
0.3120
0.0009
0.0000
tree
0.0465
0.0136
0.0509
0.8887
0.0003
0.0000
car
0.3416
0.0048
0.0003
0.0007
0.6527
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.873
0.961
0.669
0.719
0.888
--
Recall/Completeness
0.917
0.946
0.477
0.889
0.653
--
F1
0.894
0.953
0.557
0.795
0.752
--

Overall accuracy 0.908

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9473
0.0416
0.0081
0.0016
0.0014
0.0000
building
0.0305
0.9601
0.0077
0.0013
0.0005
0.0000
low_veg
0.1280
0.0105
0.5227
0.3381
0.0007
0.0000
tree
0.0188
0.0051
0.0444
0.9317
0.0000
0.0000
car
0.2247
0.0008
0.0000
0.0000
0.7745
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.911
0.978
0.701
0.748
0.924
--
Recall/Completeness
0.947
0.960
0.523
0.932
0.775
--
F1
0.929
0.969
0.599
0.830
0.843
--

Overall accuracy 0.935

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area6_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8888
0.0515
0.0440
0.0134
0.0024
0.0000
building
0.0453
0.9360
0.0142
0.0045
0.0000
0.0000
low_veg
0.0658
0.0268
0.7370
0.1704
0.0001
0.0000
tree
0.0121
0.0029
0.0442
0.9407
0.0000
0.0000
car
0.3673
0.0348
0.0013
0.0023
0.5944
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.878
0.947
0.733
0.921
0.940
--
Recall/Completeness
0.889
0.936
0.737
0.941
0.594
--
F1
0.883
0.941
0.735
0.931
0.728
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9262
0.0357
0.0334
0.0039
0.0008
0.0000
building
0.0354
0.9533
0.0098
0.0014
0.0000
0.0000
low_veg
0.0512
0.0149
0.7878
0.1462
0.0000
0.0000
tree
0.0037
0.0008
0.0336
0.9619
0.0000
0.0000
car
0.2426
0.0334
0.0001
0.0001
0.7239
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.916
0.968
0.780
0.948
0.976
--
Recall/Completeness
0.926
0.953
0.788
0.962
0.724
--
F1
0.921
0.960
0.784
0.955
0.831
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8765
0.0350
0.0738
0.0127
0.0019
0.0000
building
0.0237
0.9505
0.0206
0.0050
0.0002
0.0000
low_veg
0.0441
0.0129
0.8845
0.0585
0.0001
0.0000
tree
0.0167
0.0051
0.0645
0.9136
0.0000
0.0000
car
0.3277
0.0236
0.0112
0.0060
0.6314
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.921
0.935
0.800
0.929
0.943
--
Recall/Completeness
0.877
0.951
0.884
0.914
0.631
--
F1
0.898
0.943
0.840
0.921
0.756
--

Overall accuracy 0.900

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9057
0.0286
0.0622
0.0031
0.0005
0.0000
building
0.0145
0.9718
0.0114
0.0021
0.0002
0.0000
low_veg
0.0308
0.0057
0.9191
0.0444
0.0000
0.0000
tree
0.0059
0.0022
0.0454
0.9465
0.0000
0.0000
car
0.2099
0.0256
0.0058
0.0042
0.7545
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.954
0.955
0.843
0.957
0.980
--
Recall/Completeness
0.906
0.972
0.919
0.946
0.755
--
F1
0.929
0.963
0.880
0.952
0.853
--

Overall accuracy 0.932

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8401
0.0758
0.0562
0.0263
0.0016
0.0000
building
0.0170
0.9541
0.0155
0.0134
0.0000
0.0000
low_veg
0.0499
0.0299
0.7461
0.1742
0.0000
0.0000
tree
0.0090
0.0067
0.0615
0.9228
0.0000
0.0000
car
0.2832
0.0284
0.0016
0.0080
0.6788
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.950
0.865
0.778
0.813
0.892
--
Recall/Completeness
0.840
0.954
0.746
0.923
0.679
--
F1
0.892
0.907
0.762
0.864
0.771
--

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
0.8767
0.0625
0.0482
0.0121
0.0006
0.0000
building
0.0131
0.9672
0.0098
0.0099
0.0000
0.0000
low_veg
0.0471
0.0181
0.7922
0.1427
0.0000
0.0000
tree
0.0033
0.0028
0.0456
0.9483
0.0000
0.0000
car
0.1915
0.0279
0.0000
0.0022
0.7784
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.963
0.894
0.817
0.859
0.948
--
Recall/Completeness
0.877
0.967
0.792
0.948
0.778
--
F1
0.918
0.929
0.804
0.901
0.855
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8687
0.0310
0.0555
0.0423
0.0025
0.0000
building
0.1548
0.8289
0.0098
0.0066
0.0000
0.0000
low_veg
0.0353
0.0272
0.7529
0.1844
0.0001
0.0000
tree
0.0065
0.0060
0.0888
0.8987
0.0000
0.0000
car
0.1776
0.0025
0.0136
0.0161
0.7901
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.741
0.916
0.836
0.845
0.913
--
Recall/Completeness
0.869
0.829
0.753
0.899
0.790
--
F1
0.800
0.870
0.793
0.871
0.847
--

Overall accuracy 0.838

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
0.9014
0.0272
0.0476
0.0236
0.0003
0.0000
building
0.1634
0.8319
0.0026
0.0021
0.0000
0.0000
low_veg
0.0337
0.0185
0.7888
0.1590
0.0000
0.0000
tree
0.0024
0.0021
0.0739
0.9217
0.0000
0.0000
car
0.0807
0.0035
0.0110
0.0079
0.8969
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.752
0.942
0.865
0.875
0.987
--
Recall/Completeness
0.901
0.832
0.789
0.922
0.897
--
F1
0.820
0.884
0.825
0.898
0.940
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8442
0.0367
0.0712
0.0425
0.0055
0.0000
building
0.0068
0.9639
0.0235
0.0058
0.0000
0.0000
low_veg
0.0472
0.0485
0.7377
0.1666
0.0001
0.0000
tree
0.0190
0.0117
0.1014
0.8679
0.0000
0.0000
car
0.1581
0.0014
0.0000
0.0087
0.8318
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.906
0.873
0.775
0.845
0.634
--
Recall/Completeness
0.844
0.964
0.738
0.868
0.832
--
F1
0.874
0.916
0.756
0.857
0.719
--

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_area14_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8668
0.0318
0.0662
0.0301
0.0050
0.0000
building
0.0032
0.9827
0.0125
0.0016
0.0000
0.0000
low_veg
0.0421
0.0349
0.7701
0.1529
0.0000
0.0000
tree
0.0127
0.0056
0.0897
0.8920
0.0000
0.0000
car
0.0474
0.0009
0.0000
0.0049
0.9468
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.929
0.905
0.800
0.869
0.613
--
Recall/Completeness
0.867
0.983
0.770
0.892
0.947
--
F1
0.897
0.942
0.785
0.881
0.744
--

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7854
0.0431
0.1042
0.0651
0.0022
0.0000
building
0.0114
0.9559
0.0193
0.0134
0.0000
0.0000
low_veg
0.0112
0.0196
0.7816
0.1875
0.0002
0.0000
tree
0.0135
0.0050
0.0880
0.8935
0.0000
0.0000
car
0.1742
0.0035
0.0008
0.0187
0.8028
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.919
0.922
0.817
0.828
0.819
--
Recall/Completeness
0.785
0.956
0.782
0.893
0.803
--
F1
0.847
0.939
0.799
0.859
0.811
--

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_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8250
0.0381
0.0947
0.0412
0.0010
0.0000
building
0.0094
0.9768
0.0087
0.0051
0.0000
0.0000
low_veg
0.0085
0.0135
0.8053
0.1725
0.0002
0.0000
tree
0.0082
0.0016
0.0756
0.9146
0.0000
0.0000
car
0.0795
0.0014
0.0000
0.0096
0.9095
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.944
0.944
0.846
0.852
0.882
--
Recall/Completeness
0.825
0.977
0.805
0.915
0.909
--
F1
0.881
0.960
0.825
0.882
0.895
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8336
0.0409
0.0603
0.0620
0.0033
0.0000
building
0.0126
0.9506
0.0296
0.0072
0.0000
0.0000
low_veg
0.0276
0.0250
0.7515
0.1958
0.0000
0.0000
tree
0.0046
0.0015
0.0392
0.9548
0.0000
0.0000
car
0.1593
0.0814
0.0395
0.0202
0.6996
0.0000
clutter
0.0060
0.9914
0.0020
0.0006
0.0000
0.0000
Precision/Correctness
0.943
0.792
0.814
0.884
0.815
-nan
Recall/Completeness
0.834
0.951
0.752
0.955
0.700
0.000
F1
0.885
0.864
0.781
0.918
0.753
-nan

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8725
0.0297
0.0556
0.0402
0.0020
0.0000
building
0.0111
0.9624
0.0245
0.0020
0.0000
0.0000
low_veg
0.0228
0.0146
0.8008
0.1618
0.0000
0.0000
tree
0.0015
0.0003
0.0290
0.9691
0.0000
0.0000
car
0.0682
0.0963
0.0348
0.0094
0.7914
0.0000
clutter
0.0058
0.9937
0.0004
0.0000
0.0000
0.0000
Precision/Correctness
0.961
0.812
0.843
0.913
0.851
-nan
Recall/Completeness
0.873
0.962
0.801
0.969
0.791
0.000
F1
0.915
0.881
0.821
0.940
0.820
-nan

Overall accuracy 0.890

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
0.8809
0.0315
0.0407
0.0425
0.0044
0.0000
building
0.0144
0.9672
0.0099
0.0083
0.0001
0.0000
low_veg
0.0686
0.0326
0.7340
0.1648
0.0000
0.0000
tree
0.0140
0.0035
0.0445
0.9379
0.0000
0.0000
car
0.2625
0.0325
0.0140
0.0170
0.6740
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.923
0.905
0.842
0.848
0.838
--
Recall/Completeness
0.881
0.967
0.734
0.938
0.674
--
F1
0.902
0.935
0.784
0.891
0.747
--

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_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9161
0.0233
0.0351
0.0228
0.0026
0.0000
building
0.0098
0.9825
0.0037
0.0039
0.0000
0.0000
low_veg
0.0649
0.0193
0.7839
0.1319
0.0000
0.0000
tree
0.0042
0.0010
0.0305
0.9643
0.0000
0.0000
car
0.1578
0.0373
0.0117
0.0084
0.7848
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.946
0.937
0.880
0.891
0.882
--
Recall/Completeness
0.916
0.983
0.784
0.964
0.785
--
F1
0.931
0.959
0.829
0.926
0.831
--

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_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7914
0.0478
0.1058
0.0504
0.0045
0.0000
building
0.0510
0.8857
0.0528
0.0105
0.0000
0.0000
low_veg
0.0135
0.0338
0.8485
0.1040
0.0002
0.0000
tree
0.0070
0.0060
0.1238
0.8632
0.0000
0.0000
car
0.2524
0.0000
0.0016
0.0141
0.7318
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.866
0.882
0.860
0.808
0.792
--
Recall/Completeness
0.791
0.886
0.848
0.863
0.732
--
F1
0.827
0.884
0.854
0.835
0.761
--

Overall accuracy 0.850

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
0.8285
0.0353
0.0962
0.0365
0.0035
0.0000
building
0.0512
0.8980
0.0432
0.0076
0.0000
0.0000
low_veg
0.0112
0.0306
0.8664
0.0915
0.0002
0.0000
tree
0.0030
0.0025
0.1031
0.8914
0.0000
0.0000
car
0.1350
0.0000
0.0000
0.0087
0.8563
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.884
0.903
0.883
0.832
0.802
--
Recall/Completeness
0.829
0.898
0.866
0.891
0.856
--
F1
0.855
0.900
0.875
0.861
0.828
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9384
0.0240
0.0235
0.0088
0.0053
0.0000
building
0.0598
0.9178
0.0153
0.0066
0.0005
0.0000
low_veg
0.0962
0.0226
0.6564
0.2246
0.0001
0.0000
tree
0.0100
0.0044
0.0293
0.9562
0.0001
0.0000
car
0.3020
0.0131
0.0037
0.0076
0.6735
0.0000
clutter
0.7870
0.0264
0.1228
0.0051
0.0588
0.0000
Precision/Correctness
0.851
0.961
0.888
0.660
0.914
-nan
Recall/Completeness
0.938
0.918
0.656
0.956
0.674
0.000
F1
0.893
0.939
0.755
0.781
0.776
-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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9643
0.0143
0.0167
0.0034
0.0013
0.0000
building
0.0477
0.9380
0.0104
0.0034
0.0004
0.0000
low_veg
0.0785
0.0106
0.6991
0.2116
0.0001
0.0000
tree
0.0039
0.0012
0.0175
0.9774
0.0000
0.0000
car
0.1920
0.0134
0.0026
0.0054
0.7866
0.0000
clutter
0.8082
0.0178
0.1061
0.0024
0.0656
0.0000
Precision/Correctness
0.887
0.980
0.923
0.692
0.965
-nan
Recall/Completeness
0.964
0.938
0.699
0.977
0.787
0.000
F1
0.924
0.958
0.795
0.810
0.867
-nan

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9359
0.0205
0.0237
0.0156
0.0042
0.0000
building
0.0171
0.9650
0.0107
0.0072
0.0000
0.0000
low_veg
0.0575
0.0239
0.5992
0.3193
0.0001
0.0000
tree
0.0024
0.0015
0.0174
0.9787
0.0000
0.0000
car
0.2896
0.0091
0.0047
0.0111
0.6856
0.0000
clutter
0.5779
0.3532
0.0278
0.0208
0.0203
0.0000
Precision/Correctness
0.920
0.919
0.849
0.654
0.792
-nan
Recall/Completeness
0.936
0.965
0.599
0.979
0.686
0.000
F1
0.928
0.942
0.703
0.784
0.735
-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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9585
0.0138
0.0173
0.0072
0.0032
0.0000
building
0.0118
0.9785
0.0060
0.0037
0.0000
0.0000
low_veg
0.0365
0.0114
0.6471
0.3050
0.0000
0.0000
tree
0.0004
0.0003
0.0120
0.9874
0.0000
0.0000
car
0.1887
0.0069
0.0055
0.0077
0.7912
0.0000
clutter
0.5640
0.3774
0.0258
0.0107
0.0221
0.0000
Precision/Correctness
0.942
0.939
0.888
0.694
0.804
-nan
Recall/Completeness
0.959
0.978
0.647
0.987
0.791
0.000
F1
0.950
0.958
0.749
0.815
0.797
-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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9537
0.0261
0.0126
0.0048
0.0028
0.0000
building
0.1713
0.8132
0.0110
0.0034
0.0012
0.0000
low_veg
0.0411
0.0061
0.9276
0.0253
0.0000
0.0000
tree
0.0300
0.0030
0.0862
0.8807
0.0002
0.0000
car
0.1821
0.2962
0.0052
0.0096
0.5070
0.0000
clutter
0.5092
0.4323
0.0307
0.0096
0.0183
0.0000
Precision/Correctness
0.821
0.860
0.880
0.933
0.899
-nan
Recall/Completeness
0.954
0.813
0.928
0.881
0.507
0.000
F1
0.883
0.836
0.903
0.906
0.648
-nan

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9689
0.0201
0.0091
0.0008
0.0011
0.0000
building
0.1657
0.8270
0.0053
0.0010
0.0010
0.0000
low_veg
0.0230
0.0036
0.9557
0.0177
0.0000
0.0000
tree
0.0100
0.0006
0.0685
0.9209
0.0000
0.0000
car
0.0975
0.3300
0.0011
0.0043
0.5671
0.0000
clutter
0.4626
0.4875
0.0256
0.0030
0.0213
0.0000
Precision/Correctness
0.845
0.880
0.914
0.967
0.923
-nan
Recall/Completeness
0.969
0.827
0.956
0.921
0.567
0.000
F1
0.903
0.853
0.935
0.943
0.703
-nan

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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9551
0.0291
0.0118
0.0032
0.0007
0.0000
building
0.1508
0.7536
0.0777
0.0165
0.0014
0.0000
low_veg
0.0721
0.0196
0.7734
0.1349
0.0000
0.0000
tree
0.0330
0.0090
0.1443
0.8137
0.0000
0.0000
car
0.3410
0.0638
0.0111
0.0016
0.5825
0.0000
clutter
0.7265
0.2460
0.0029
0.0003
0.0242
0.0000
Precision/Correctness
0.783
0.882
0.737
0.761
0.789
-nan
Recall/Completeness
0.955
0.754
0.773
0.814
0.582
0.000
F1
0.861
0.813
0.755
0.786
0.670
-nan

Overall accuracy 0.798

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
0.9698
0.0230
0.0063
0.0005
0.0005
0.0000
building
0.1478
0.7603
0.0760
0.0146
0.0013
0.0000
low_veg
0.0431
0.0130
0.8024
0.1415
0.0000
0.0000
tree
0.0118
0.0052
0.1313
0.8517
0.0000
0.0000
car
0.2242
0.0760
0.0064
0.0003
0.6932
0.0000
clutter
0.7112
0.2605
0.0014
0.0000
0.0269
0.0000
Precision/Correctness
0.806
0.901
0.758
0.770
0.782
-nan
Recall/Completeness
0.970
0.760
0.802
0.852
0.693
0.000
F1
0.880
0.825
0.779
0.809
0.735
-nan

Overall accuracy 0.818

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
0.8470
0.0377
0.0934
0.0212
0.0007
0.0000
building
0.0330
0.9404
0.0198
0.0063
0.0004
0.0000
low_veg
0.0319
0.0094
0.7338
0.2250
0.0000
0.0000
tree
0.0201
0.0044
0.1113
0.8642
0.0000
0.0000
car
0.3472
0.0025
0.0001
0.0103
0.6398
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.898
0.951
0.769
0.763
0.940
--
Recall/Completeness
0.847
0.940
0.734
0.864
0.640
--
F1
0.872
0.946
0.751
0.810
0.761
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8987
0.0323
0.0639
0.0050
0.0002
0.0000
building
0.0235
0.9667
0.0077
0.0017
0.0004
0.0000
low_veg
0.0205
0.0031
0.7591
0.2173
0.0000
0.0000
tree
0.0056
0.0007
0.0829
0.9108
0.0000
0.0000
car
0.2231
0.0020
0.0000
0.0032
0.7717
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.940
0.968
0.835
0.791
0.966
--
Recall/Completeness
0.899
0.967
0.759
0.911
0.772
--
F1
0.919
0.968
0.795
0.847
0.858
--

Overall accuracy 0.886

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8948
0.0256
0.0607
0.0172
0.0018
0.0000
building
0.0220
0.9189
0.0499
0.0089
0.0003
0.0000
low_veg
0.0375
0.0230
0.8434
0.0961
0.0000
0.0000
tree
0.0145
0.0029
0.1106
0.8720
0.0000
0.0000
car
0.2745
0.0298
0.0384
0.0165
0.6408
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.918
0.943
0.823
0.834
0.916
--
Recall/Completeness
0.895
0.919
0.843
0.872
0.641
--
F1
0.906
0.931
0.833
0.852
0.754
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9265
0.0225
0.0451
0.0054
0.0005
0.0000
building
0.0147
0.9446
0.0355
0.0050
0.0003
0.0000
low_veg
0.0235
0.0196
0.8775
0.0795
0.0000
0.0000
tree
0.0048
0.0007
0.0876
0.9069
0.0000
0.0000
car
0.1684
0.0347
0.0325
0.0111
0.7533
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.952
0.955
0.864
0.874
0.964
--
Recall/Completeness
0.926
0.945
0.878
0.907
0.753
--
F1
0.939
0.950
0.870
0.890
0.846
--

Overall accuracy 0.913

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image