Detailled semantic labeling (2D) result


Name D. Marmanis et al.
Affiliation DLR-DFD, Germany
Abbreviation DLR_8
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
0.911
0.026
0.044
0.016
0.003
0.000
building
0.073
0.899
0.021
0.006
0.000
0.000
low_veg
0.057
0.021
0.796
0.126
0.000
0.000
tree
0.020
0.006
0.097
0.877
0.000
0.000
car
0.367
0.053
0.010
0.008
0.562
0.000
clutter
0.697
0.253
0.021
0.010
0.019
0.000
Precision/Correctness
0.845
0.936
0.811
0.856
0.878
-nan
Recall/Completeness
0.911
0.899
0.796
0.877
0.562
0.000
F1
0.877
0.917
0.803
0.867
0.685
-nan

Overall accuracy 0.863

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
0.936
0.019
0.036
0.008
0.001
0.000
building
0.065
0.916
0.015
0.003
0.000
0.000
low_veg
0.045
0.012
0.832
0.111
0.000
0.000
tree
0.009
0.002
0.080
0.909
0.000
0.000
car
0.258
0.060
0.008
0.006
0.668
0.000
clutter
0.690
0.268
0.017
0.005
0.020
0.000
Precision/Correctness
0.874
0.957
0.845
0.885
0.906
-nan
Recall/Completeness
0.936
0.916
0.832
0.909
0.668
0.000
F1
0.904
0.936
0.839
0.897
0.769
-nan

Overall accuracy 0.892
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9526
0.0086
0.0249
0.0108
0.0031
0.0000
building
0.0377
0.9372
0.0216
0.0035
0.0000
0.0000
low_veg
0.0813
0.0138
0.7553
0.1496
0.0000
0.0000
tree
0.0300
0.0054
0.0886
0.8760
0.0000
0.0000
car
0.3869
0.0261
0.0011
0.0021
0.5838
0.0000
clutter
0.9806
0.0000
0.0000
0.0194
0.0000
0.0000
Precision/Correctness
0.798
0.967
0.848
0.842
0.920
-nan
Recall/Completeness
0.953
0.937
0.755
0.876
0.584
0.000
F1
0.869
0.952
0.799
0.859
0.714
-nan

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9749
0.0052
0.0145
0.0032
0.0023
0.0000
building
0.0292
0.9577
0.0121
0.0010
0.0000
0.0000
low_veg
0.0636
0.0055
0.7883
0.1427
0.0000
0.0000
tree
0.0144
0.0021
0.0748
0.9087
0.0000
0.0000
car
0.2622
0.0300
0.0003
0.0004
0.7071
0.0000
clutter
0.9962
0.0000
0.0000
0.0038
0.0000
0.0000
Precision/Correctness
0.832
0.985
0.883
0.865
0.938
-nan
Recall/Completeness
0.975
0.958
0.788
0.909
0.707
0.000
F1
0.898
0.971
0.833
0.886
0.806
-nan

Overall accuracy 0.887

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
0.9524
0.0347
0.0076
0.0028
0.0024
0.0000
building
0.0693
0.9173
0.0106
0.0021
0.0006
0.0000
low_veg
0.1531
0.0433
0.5315
0.2718
0.0003
0.0000
tree
0.0585
0.0181
0.0760
0.8474
0.0000
0.0000
car
0.4112
0.0035
0.0007
0.0026
0.5821
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.836
0.975
0.667
0.742
0.888
--
Recall/Completeness
0.952
0.917
0.532
0.847
0.582
--
F1
0.890
0.945
0.591
0.791
0.703
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9758
0.0169
0.0053
0.0009
0.0010
0.0000
building
0.0538
0.9349
0.0095
0.0011
0.0006
0.0000
low_veg
0.1051
0.0121
0.5829
0.2995
0.0004
0.0000
tree
0.0249
0.0063
0.0707
0.8982
0.0000
0.0000
car
0.2931
0.0003
0.0001
0.0031
0.7035
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.873
0.990
0.691
0.765
0.922
--
Recall/Completeness
0.976
0.935
0.583
0.898
0.703
--
F1
0.922
0.962
0.632
0.826
0.798
--

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_area6_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8985
0.0419
0.0484
0.0084
0.0028
0.0000
building
0.0627
0.9150
0.0171
0.0042
0.0009
0.0000
low_veg
0.0785
0.0250
0.7502
0.1463
0.0000
0.0000
tree
0.0172
0.0047
0.0915
0.8866
0.0000
0.0000
car
0.4064
0.0041
0.0006
0.0009
0.5880
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.849
0.953
0.643
0.930
0.905
--
Recall/Completeness
0.898
0.915
0.750
0.887
0.588
--
F1
0.873
0.934
0.693
0.908
0.713
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9299
0.0282
0.0385
0.0024
0.0010
0.0000
building
0.0525
0.9329
0.0125
0.0014
0.0008
0.0000
low_veg
0.0649
0.0100
0.7992
0.1260
0.0000
0.0000
tree
0.0064
0.0020
0.0818
0.9098
0.0000
0.0000
car
0.2811
0.0005
0.0001
0.0000
0.7183
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.886
0.974
0.671
0.953
0.944
--
Recall/Completeness
0.930
0.933
0.799
0.910
0.718
--
F1
0.908
0.953
0.729
0.931
0.816
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8918
0.0308
0.0663
0.0095
0.0015
0.0000
building
0.0421
0.9349
0.0182
0.0047
0.0002
0.0000
low_veg
0.0515
0.0117
0.8913
0.0454
0.0002
0.0000
tree
0.0231
0.0071
0.0794
0.8905
0.0000
0.0000
car
0.3898
0.0071
0.0118
0.0042
0.5871
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.897
0.940
0.799
0.942
0.948
--
Recall/Completeness
0.892
0.935
0.891
0.891
0.587
--
F1
0.895
0.937
0.842
0.916
0.725
--

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
0.9165
0.0250
0.0556
0.0024
0.0004
0.0000
building
0.0309
0.9572
0.0096
0.0022
0.0001
0.0000
low_veg
0.0350
0.0048
0.9240
0.0361
0.0001
0.0000
tree
0.0084
0.0029
0.0583
0.9304
0.0000
0.0000
car
0.2705
0.0058
0.0043
0.0025
0.7168
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.936
0.960
0.842
0.964
0.982
--
Recall/Completeness
0.917
0.957
0.924
0.930
0.717
--
F1
0.926
0.959
0.881
0.947
0.829
--

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
0.8649
0.0456
0.0652
0.0223
0.0021
0.0000
building
0.0290
0.9434
0.0123
0.0148
0.0004
0.0000
low_veg
0.0729
0.0286
0.7264
0.1721
0.0000
0.0000
tree
0.0139
0.0073
0.0684
0.9104
0.0000
0.0000
car
0.3772
0.0187
0.0022
0.0081
0.5938
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.928
0.904
0.755
0.816
0.823
--
Recall/Completeness
0.865
0.943
0.726
0.910
0.594
--
F1
0.895
0.923
0.741
0.861
0.690
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8961
0.0335
0.0587
0.0104
0.0013
0.0000
building
0.0239
0.9574
0.0070
0.0114
0.0003
0.0000
low_veg
0.0719
0.0153
0.7687
0.1441
0.0000
0.0000
tree
0.0047
0.0032
0.0515
0.9406
0.0000
0.0000
car
0.2951
0.0215
0.0000
0.0021
0.6813
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.944
0.934
0.790
0.858
0.860
--
Recall/Completeness
0.896
0.957
0.769
0.941
0.681
--
F1
0.919
0.946
0.779
0.897
0.760
--

Overall accuracy 0.897

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8864
0.0278
0.0503
0.0339
0.0016
0.0000
building
0.1473
0.8391
0.0083
0.0054
0.0000
0.0000
low_veg
0.0472
0.0275
0.7563
0.1688
0.0002
0.0000
tree
0.0091
0.0070
0.1047
0.8791
0.0000
0.0000
car
0.2471
0.0002
0.0092
0.0094
0.7341
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.731
0.917
0.822
0.856
0.930
--
Recall/Completeness
0.886
0.839
0.756
0.879
0.734
--
F1
0.801
0.876
0.788
0.867
0.821
--

Overall accuracy 0.836

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
0.9146
0.0237
0.0427
0.0189
0.0001
0.0000
building
0.1536
0.8424
0.0023
0.0017
0.0000
0.0000
low_veg
0.0452
0.0185
0.7916
0.1446
0.0001
0.0000
tree
0.0037
0.0025
0.0893
0.9045
0.0000
0.0000
car
0.1299
0.0002
0.0071
0.0029
0.8599
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.746
0.945
0.849
0.884
0.988
--
Recall/Completeness
0.915
0.842
0.792
0.904
0.860
--
F1
0.822
0.891
0.819
0.894
0.920
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8429
0.0306
0.0893
0.0323
0.0049
0.0000
building
0.0114
0.9646
0.0192
0.0048
0.0000
0.0000
low_veg
0.0561
0.0469
0.7620
0.1350
0.0001
0.0000
tree
0.0213
0.0130
0.1126
0.8531
0.0000
0.0000
car
0.1796
0.0000
0.0000
0.0310
0.7894
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.890
0.879
0.759
0.870
0.649
--
Recall/Completeness
0.843
0.965
0.762
0.853
0.789
--
F1
0.866
0.920
0.760
0.862
0.713
--

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_area14_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8613
0.0266
0.0851
0.0225
0.0045
0.0000
building
0.0071
0.9815
0.0101
0.0013
0.0000
0.0000
low_veg
0.0507
0.0324
0.7950
0.1218
0.0000
0.0000
tree
0.0138
0.0061
0.1001
0.8800
0.0000
0.0000
car
0.0597
0.0000
0.0000
0.0355
0.9049
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.915
0.913
0.781
0.893
0.630
--
Recall/Completeness
0.861
0.981
0.795
0.880
0.905
--
F1
0.887
0.946
0.788
0.886
0.743
--

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
0.8219
0.0267
0.0968
0.0523
0.0023
0.0000
building
0.0192
0.9546
0.0139
0.0116
0.0007
0.0000
low_veg
0.0194
0.0153
0.8554
0.1097
0.0002
0.0000
tree
0.0183
0.0056
0.1091
0.8670
0.0000
0.0000
car
0.2743
0.0023
0.0024
0.0174
0.7036
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.885
0.941
0.815
0.882
0.751
--
Recall/Completeness
0.822
0.955
0.855
0.867
0.704
--
F1
0.852
0.947
0.835
0.874
0.727
--

Overall accuracy 0.872

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8571
0.0221
0.0872
0.0326
0.0010
0.0000
building
0.0162
0.9739
0.0047
0.0046
0.0007
0.0000
low_veg
0.0160
0.0089
0.8814
0.0935
0.0001
0.0000
tree
0.0116
0.0016
0.0950
0.8918
0.0000
0.0000
car
0.2064
0.0000
0.0000
0.0039
0.7897
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.913
0.964
0.843
0.907
0.802
--
Recall/Completeness
0.857
0.974
0.881
0.892
0.790
--
F1
0.884
0.969
0.862
0.899
0.796
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8703
0.0281
0.0561
0.0429
0.0025
0.0000
building
0.0335
0.9311
0.0303
0.0051
0.0000
0.0000
low_veg
0.0350
0.0273
0.7580
0.1797
0.0000
0.0000
tree
0.0068
0.0019
0.0468
0.9445
0.0000
0.0000
car
0.2594
0.0449
0.0186
0.0124
0.6648
0.0000
clutter
0.4994
0.4991
0.0007
0.0008
0.0000
0.0000
Precision/Correctness
0.855
0.857
0.805
0.897
0.843
-nan
Recall/Completeness
0.870
0.931
0.758
0.944
0.665
0.000
F1
0.863
0.893
0.781
0.920
0.743
-nan

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9031
0.0193
0.0527
0.0234
0.0015
0.0000
building
0.0321
0.9417
0.0252
0.0010
0.0000
0.0000
low_veg
0.0274
0.0173
0.8055
0.1497
0.0000
0.0000
tree
0.0026
0.0005
0.0352
0.9617
0.0000
0.0000
car
0.1680
0.0499
0.0081
0.0036
0.7704
0.0000
clutter
0.5070
0.4928
0.0002
0.0000
0.0000
0.0000
Precision/Correctness
0.873
0.880
0.834
0.925
0.884
-nan
Recall/Completeness
0.903
0.942
0.805
0.962
0.770
0.000
F1
0.888
0.910
0.820
0.943
0.823
-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
0.8930
0.0285
0.0430
0.0311
0.0044
0.0000
building
0.0318
0.9526
0.0089
0.0066
0.0001
0.0000
low_veg
0.0781
0.0349
0.7429
0.1440
0.0000
0.0000
tree
0.0197
0.0045
0.0621
0.9138
0.0000
0.0000
car
0.3271
0.0136
0.0112
0.0079
0.6402
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.903
0.906
0.818
0.867
0.829
--
Recall/Completeness
0.893
0.953
0.743
0.914
0.640
--
F1
0.898
0.929
0.779
0.890
0.722
--

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_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9235
0.0216
0.0370
0.0153
0.0026
0.0000
building
0.0267
0.9677
0.0024
0.0032
0.0000
0.0000
low_veg
0.0741
0.0221
0.7892
0.1146
0.0000
0.0000
tree
0.0071
0.0013
0.0454
0.9462
0.0000
0.0000
car
0.2215
0.0156
0.0104
0.0015
0.7510
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.928
0.937
0.858
0.907
0.876
--
Recall/Completeness
0.924
0.968
0.789
0.946
0.751
--
F1
0.926
0.952
0.822
0.926
0.809
--

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_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8280
0.0364
0.0938
0.0364
0.0053
0.0000
building
0.0685
0.8842
0.0319
0.0154
0.0000
0.0000
low_veg
0.0348
0.0239
0.8739
0.0674
0.0000
0.0000
tree
0.0143
0.0068
0.1652
0.8137
0.0000
0.0000
car
0.2917
0.0000
0.0022
0.0060
0.7002
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.791
0.908
0.854
0.852
0.774
--
Recall/Completeness
0.828
0.884
0.874
0.814
0.700
--
F1
0.809
0.896
0.864
0.832
0.735
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8603
0.0255
0.0850
0.0251
0.0041
0.0000
building
0.0665
0.8963
0.0243
0.0129
0.0000
0.0000
low_veg
0.0316
0.0201
0.8910
0.0572
0.0000
0.0000
tree
0.0085
0.0029
0.1427
0.8458
0.0000
0.0000
car
0.1810
0.0000
0.0000
0.0014
0.8176
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.812
0.930
0.877
0.876
0.790
--
Recall/Completeness
0.860
0.896
0.891
0.846
0.818
--
F1
0.835
0.913
0.884
0.860
0.804
--

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
0.9497
0.0202
0.0202
0.0066
0.0033
0.0000
building
0.0757
0.9033
0.0154
0.0050
0.0005
0.0000
low_veg
0.0882
0.0230
0.6986
0.1901
0.0001
0.0000
tree
0.0120
0.0048
0.0400
0.9431
0.0000
0.0000
car
0.4511
0.0120
0.0063
0.0141
0.5165
0.0000
clutter
0.7974
0.0404
0.0967
0.0027
0.0628
0.0000
Precision/Correctness
0.831
0.964
0.894
0.695
0.924
-nan
Recall/Completeness
0.950
0.903
0.699
0.943
0.516
0.000
F1
0.887
0.932
0.784
0.800
0.663
-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
0.9713
0.0116
0.0137
0.0025
0.0008
0.0000
building
0.0634
0.9242
0.0098
0.0023
0.0004
0.0000
low_veg
0.0685
0.0099
0.7437
0.1778
0.0000
0.0000
tree
0.0048
0.0016
0.0256
0.9680
0.0000
0.0000
car
0.3453
0.0120
0.0054
0.0142
0.6231
0.0000
clutter
0.8098
0.0387
0.0814
0.0002
0.0699
0.0000
Precision/Correctness
0.869
0.982
0.929
0.726
0.966
-nan
Recall/Completeness
0.971
0.924
0.744
0.968
0.623
0.000
F1
0.917
0.952
0.826
0.830
0.757
-nan

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9459
0.0159
0.0231
0.0123
0.0028
0.0000
building
0.0642
0.9189
0.0100
0.0069
0.0000
0.0000
low_veg
0.0857
0.0234
0.6143
0.2765
0.0002
0.0000
tree
0.0047
0.0019
0.0247
0.9686
0.0000
0.0000
car
0.3638
0.0022
0.0057
0.0090
0.6193
0.0000
clutter
0.5731
0.3114
0.0584
0.0174
0.0396
0.0000
Precision/Correctness
0.884
0.927
0.842
0.686
0.777
-nan
Recall/Completeness
0.946
0.919
0.614
0.969
0.619
0.000
F1
0.914
0.923
0.711
0.803
0.689
-nan

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9656
0.0100
0.0168
0.0056
0.0021
0.0000
building
0.0587
0.9325
0.0052
0.0036
0.0000
0.0000
low_veg
0.0644
0.0101
0.6604
0.2651
0.0000
0.0000
tree
0.0011
0.0004
0.0184
0.9801
0.0000
0.0000
car
0.2559
0.0001
0.0065
0.0057
0.7318
0.0000
clutter
0.5586
0.3309
0.0581
0.0080
0.0444
0.0000
Precision/Correctness
0.906
0.947
0.880
0.723
0.780
-nan
Recall/Completeness
0.966
0.933
0.660
0.980
0.732
0.000
F1
0.935
0.940
0.755
0.832
0.755
-nan

Overall accuracy 0.887

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
0.9632
0.0157
0.0156
0.0035
0.0020
0.0000
building
0.2468
0.7397
0.0099
0.0031
0.0005
0.0000
low_veg
0.0395
0.0039
0.9391
0.0175
0.0000
0.0000
tree
0.0406
0.0036
0.0925
0.8632
0.0001
0.0000
car
0.2570
0.2825
0.0108
0.0092
0.4404
0.0000
clutter
0.5592
0.3821
0.0329
0.0134
0.0125
0.0000
Precision/Correctness
0.778
0.872
0.871
0.946
0.922
-nan
Recall/Completeness
0.963
0.740
0.939
0.863
0.440
0.000
F1
0.861
0.800
0.904
0.903
0.596
-nan

Overall accuracy 0.835

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
0.9762
0.0104
0.0120
0.0006
0.0008
0.0000
building
0.2413
0.7533
0.0045
0.0005
0.0004
0.0000
low_veg
0.0225
0.0019
0.9642
0.0113
0.0000
0.0000
tree
0.0140
0.0007
0.0740
0.9112
0.0000
0.0000
car
0.1592
0.3161
0.0057
0.0056
0.5135
0.0000
clutter
0.5147
0.4346
0.0272
0.0086
0.0149
0.0000
Precision/Correctness
0.800
0.891
0.906
0.976
0.943
-nan
Recall/Completeness
0.976
0.753
0.964
0.911
0.513
0.000
F1
0.879
0.817
0.934
0.942
0.665
-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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9805
0.0079
0.0088
0.0020
0.0009
0.0000
building
0.2228
0.6793
0.0781
0.0180
0.0018
0.0000
low_veg
0.0982
0.0051
0.8664
0.0296
0.0007
0.0000
tree
0.0468
0.0025
0.1678
0.7830
0.0000
0.0000
car
0.4532
0.0018
0.0072
0.0154
0.5224
0.0000
clutter
0.8255
0.1414
0.0028
0.0039
0.0265
0.0000
Precision/Correctness
0.737
0.944
0.753
0.882
0.736
-nan
Recall/Completeness
0.980
0.679
0.866
0.783
0.522
0.000
F1
0.841
0.790
0.806
0.830
0.611
-nan

Overall accuracy 0.796

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
0.9902
0.0048
0.0043
0.0003
0.0006
0.0000
building
0.2185
0.6875
0.0759
0.0164
0.0017
0.0000
low_veg
0.0667
0.0010
0.9034
0.0281
0.0008
0.0000
tree
0.0201
0.0004
0.1519
0.8275
0.0000
0.0000
car
0.3537
0.0000
0.0031
0.0194
0.6238
0.0000
clutter
0.8150
0.1501
0.0013
0.0040
0.0296
0.0000
Precision/Correctness
0.757
0.956
0.773
0.897
0.726
-nan
Recall/Completeness
0.990
0.687
0.903
0.828
0.624
0.000
F1
0.858
0.800
0.833
0.861
0.671
-nan

Overall accuracy 0.816

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
0.8656
0.0302
0.0895
0.0138
0.0009
0.0000
building
0.0385
0.9370
0.0199
0.0043
0.0004
0.0000
low_veg
0.0361
0.0117
0.7653
0.1870
0.0000
0.0000
tree
0.0277
0.0058
0.1382
0.8284
0.0000
0.0000
car
0.3691
0.0015
0.0003
0.0059
0.6231
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.884
0.955
0.760
0.792
0.937
--
Recall/Completeness
0.866
0.937
0.765
0.828
0.623
--
F1
0.875
0.946
0.762
0.810
0.748
--

Overall accuracy 0.849

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
0.9108
0.0249
0.0605
0.0035
0.0002
0.0000
building
0.0283
0.9618
0.0085
0.0010
0.0004
0.0000
low_veg
0.0232
0.0043
0.7901
0.1825
0.0000
0.0000
tree
0.0086
0.0010
0.1061
0.8843
0.0000
0.0000
car
0.2513
0.0009
0.0000
0.0010
0.7468
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.929
0.973
0.823
0.815
0.963
--
Recall/Completeness
0.911
0.962
0.790
0.884
0.747
--
F1
0.920
0.968
0.806
0.848
0.841
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9057
0.0278
0.0528
0.0117
0.0020
0.0000
building
0.0285
0.9252
0.0410
0.0049
0.0004
0.0000
low_veg
0.0501
0.0132
0.8729
0.0638
0.0000
0.0000
tree
0.0227
0.0037
0.1333
0.8402
0.0000
0.0000
car
0.3612
0.0016
0.0538
0.0052
0.5782
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.893
0.952
0.827
0.880
0.892
--
Recall/Completeness
0.906
0.925
0.873
0.840
0.578
--
F1
0.899
0.938
0.849
0.860
0.702
--

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
0.9317
0.0251
0.0385
0.0039
0.0008
0.0000
building
0.0202
0.9506
0.0268
0.0020
0.0004
0.0000
low_veg
0.0348
0.0080
0.9068
0.0504
0.0000
0.0000
tree
0.0084
0.0009
0.1058
0.8848
0.0000
0.0000
car
0.2587
0.0002
0.0531
0.0008
0.6871
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.931
0.965
0.869
0.915
0.938
--
Recall/Completeness
0.932
0.951
0.907
0.885
0.687
--
F1
0.932
0.958
0.887
0.900
0.793
--

Overall accuracy 0.919

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image