Detailled semantic labeling (2D) result


Name D. Marmanis et al.
Affiliation DLR-DFD, Germany
Abbreviation DLR_5
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
0.909
0.025
0.048
0.014
0.004
0.000
building
0.083
0.889
0.022
0.006
0.001
0.000
low_veg
0.062
0.019
0.796
0.122
0.000
0.000
tree
0.020
0.006
0.115
0.859
0.000
0.000
car
0.331
0.053
0.011
0.009
0.597
0.000
clutter
0.692
0.248
0.026
0.010
0.025
0.000
Precision/Correctness
0.835
0.938
0.791
0.858
0.832
-nan
Recall/Completeness
0.909
0.889
0.796
0.859
0.597
0.000
F1
0.871
0.913
0.793
0.858
0.695
-nan

Overall accuracy 0.856

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
0.935
0.018
0.038
0.007
0.002
0.000
building
0.075
0.905
0.015
0.004
0.001
0.000
low_veg
0.051
0.011
0.829
0.109
0.000
0.000
tree
0.009
0.002
0.097
0.892
0.000
0.000
car
0.235
0.060
0.008
0.008
0.689
0.000
clutter
0.685
0.261
0.022
0.006
0.027
0.000
Precision/Correctness
0.863
0.958
0.827
0.885
0.870
-nan
Recall/Completeness
0.935
0.905
0.829
0.892
0.689
0.000
F1
0.897
0.931
0.828
0.889
0.769
-nan

Overall accuracy 0.884
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9510
0.0082
0.0265
0.0097
0.0046
0.0000
building
0.0477
0.9263
0.0226
0.0033
0.0001
0.0000
low_veg
0.0822
0.0124
0.7581
0.1473
0.0000
0.0000
tree
0.0313
0.0056
0.0990
0.8640
0.0000
0.0000
car
0.3353
0.0304
0.0029
0.0019
0.6295
0.0000
clutter
0.9821
0.0012
0.0000
0.0167
0.0000
0.0000
Precision/Correctness
0.793
0.968
0.836
0.844
0.890
-nan
Recall/Completeness
0.951
0.926
0.758
0.864
0.629
0.000
F1
0.865
0.947
0.795
0.854
0.737
-nan

Overall accuracy 0.855

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
0.9743
0.0049
0.0150
0.0027
0.0031
0.0000
building
0.0379
0.9475
0.0132
0.0012
0.0001
0.0000
low_veg
0.0644
0.0049
0.7891
0.1416
0.0000
0.0000
tree
0.0159
0.0025
0.0839
0.8976
0.0000
0.0000
car
0.2238
0.0349
0.0018
0.0005
0.7390
0.0000
clutter
0.9960
0.0012
0.0000
0.0028
0.0000
0.0000
Precision/Correctness
0.826
0.985
0.872
0.865
0.917
-nan
Recall/Completeness
0.974
0.948
0.789
0.898
0.739
0.000
F1
0.894
0.966
0.828
0.881
0.818
-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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9542
0.0311
0.0082
0.0028
0.0037
0.0000
building
0.0796
0.9045
0.0126
0.0023
0.0010
0.0000
low_veg
0.1419
0.0364
0.5520
0.2688
0.0010
0.0000
tree
0.0544
0.0158
0.1249
0.8049
0.0000
0.0000
car
0.3677
0.0032
0.0018
0.0052
0.6221
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.825
0.978
0.613
0.732
0.844
--
Recall/Completeness
0.954
0.905
0.552
0.805
0.622
--
F1
0.885
0.940
0.581
0.767
0.716
--

Overall accuracy 0.894

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
0.9775
0.0145
0.0052
0.0011
0.0018
0.0000
building
0.0642
0.9220
0.0115
0.0014
0.0009
0.0000
low_veg
0.0973
0.0082
0.5959
0.2976
0.0010
0.0000
tree
0.0228
0.0059
0.1154
0.8559
0.0000
0.0000
car
0.2600
0.0003
0.0002
0.0050
0.7345
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.859
0.991
0.636
0.754
0.880
--
Recall/Completeness
0.977
0.922
0.596
0.856
0.735
--
F1
0.915
0.955
0.615
0.802
0.801
--

Overall accuracy 0.920

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
0.8951
0.0409
0.0522
0.0065
0.0053
0.0000
building
0.0741
0.9000
0.0203
0.0044
0.0013
0.0000
low_veg
0.0843
0.0247
0.7453
0.1456
0.0000
0.0000
tree
0.0177
0.0061
0.1402
0.8360
0.0000
0.0000
car
0.3577
0.0022
0.0008
0.0006
0.6387
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.837
0.952
0.568
0.928
0.855
--
Recall/Completeness
0.895
0.900
0.745
0.836
0.639
--
F1
0.865
0.925
0.645
0.880
0.731
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9263
0.0285
0.0410
0.0017
0.0025
0.0000
building
0.0641
0.9175
0.0151
0.0021
0.0012
0.0000
low_veg
0.0712
0.0106
0.7884
0.1298
0.0000
0.0000
tree
0.0073
0.0034
0.1304
0.8589
0.0000
0.0000
car
0.2454
0.0001
0.0002
0.0000
0.7544
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.871
0.972
0.586
0.949
0.901
--
Recall/Completeness
0.926
0.918
0.788
0.859
0.754
--
F1
0.898
0.944
0.672
0.902
0.821
--

Overall accuracy 0.888

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
0.8916
0.0294
0.0686
0.0074
0.0030
0.0000
building
0.0545
0.9212
0.0189
0.0050
0.0003
0.0000
low_veg
0.0547
0.0110
0.8870
0.0470
0.0002
0.0000
tree
0.0233
0.0065
0.0942
0.8760
0.0000
0.0000
car
0.3357
0.0087
0.0121
0.0063
0.6372
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.890
0.942
0.782
0.942
0.910
--
Recall/Completeness
0.892
0.921
0.887
0.876
0.637
--
F1
0.891
0.931
0.831
0.908
0.750
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9161
0.0242
0.0569
0.0017
0.0010
0.0000
building
0.0426
0.9438
0.0103
0.0030
0.0003
0.0000
low_veg
0.0377
0.0045
0.9180
0.0397
0.0001
0.0000
tree
0.0090
0.0025
0.0698
0.9187
0.0000
0.0000
car
0.2303
0.0068
0.0064
0.0066
0.7500
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.927
0.961
0.828
0.961
0.958
--
Recall/Completeness
0.916
0.944
0.918
0.919
0.750
--
F1
0.921
0.952
0.870
0.939
0.842
--

Overall accuracy 0.922

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
0.8636
0.0415
0.0707
0.0208
0.0035
0.0000
building
0.0350
0.9378
0.0120
0.0145
0.0006
0.0000
low_veg
0.0826
0.0257
0.7231
0.1686
0.0000
0.0000
tree
0.0135
0.0073
0.0800
0.8991
0.0000
0.0000
car
0.3376
0.0179
0.0019
0.0106
0.6319
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.920
0.911
0.735
0.819
0.754
--
Recall/Completeness
0.864
0.938
0.723
0.899
0.632
--
F1
0.891
0.924
0.729
0.857
0.688
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8949
0.0296
0.0637
0.0096
0.0021
0.0000
building
0.0288
0.9524
0.0071
0.0113
0.0005
0.0000
low_veg
0.0816
0.0122
0.7639
0.1423
0.0000
0.0000
tree
0.0049
0.0031
0.0609
0.9311
0.0000
0.0000
car
0.2758
0.0203
0.0000
0.0089
0.6950
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.937
0.942
0.771
0.859
0.795
--
Recall/Completeness
0.895
0.952
0.764
0.931
0.695
--
F1
0.915
0.947
0.767
0.894
0.742
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8831
0.0274
0.0544
0.0315
0.0036
0.0000
building
0.1454
0.8416
0.0084
0.0047
0.0000
0.0000
low_veg
0.0526
0.0268
0.7500
0.1703
0.0004
0.0000
tree
0.0095
0.0074
0.1158
0.8673
0.0000
0.0000
car
0.2145
0.0016
0.0067
0.0073
0.7700
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.725
0.918
0.806
0.854
0.861
--
Recall/Completeness
0.883
0.842
0.750
0.867
0.770
--
F1
0.796
0.878
0.777
0.861
0.813
--

Overall accuracy 0.830

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
0.9121
0.0238
0.0454
0.0180
0.0008
0.0000
building
0.1510
0.8446
0.0030
0.0014
0.0000
0.0000
low_veg
0.0501
0.0181
0.7843
0.1472
0.0003
0.0000
tree
0.0040
0.0023
0.1001
0.8936
0.0000
0.0000
car
0.1190
0.0017
0.0043
0.0024
0.8725
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.741
0.946
0.834
0.882
0.937
--
Recall/Completeness
0.912
0.845
0.784
0.894
0.873
--
F1
0.818
0.892
0.808
0.888
0.903
--

Overall accuracy 0.855

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
0.8404
0.0272
0.0956
0.0308
0.0060
0.0000
building
0.0190
0.9563
0.0207
0.0040
0.0000
0.0000
low_veg
0.0643
0.0432
0.7604
0.1320
0.0000
0.0000
tree
0.0221
0.0126
0.1284
0.8369
0.0000
0.0000
car
0.1204
0.0000
0.0050
0.0366
0.8380
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.876
0.887
0.738
0.871
0.616
--
Recall/Completeness
0.840
0.956
0.760
0.837
0.838
--
F1
0.858
0.920
0.749
0.854
0.710
--

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_area14_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8593
0.0234
0.0906
0.0215
0.0052
0.0000
building
0.0144
0.9732
0.0114
0.0010
0.0000
0.0000
low_veg
0.0591
0.0291
0.7916
0.1202
0.0000
0.0000
tree
0.0145
0.0059
0.1154
0.8642
0.0000
0.0000
car
0.0318
0.0000
0.0025
0.0429
0.9228
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.900
0.920
0.760
0.893
0.601
--
Recall/Completeness
0.859
0.973
0.792
0.864
0.923
--
F1
0.879
0.946
0.775
0.878
0.728
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8240
0.0229
0.1016
0.0485
0.0031
0.0000
building
0.0285
0.9449
0.0161
0.0097
0.0009
0.0000
low_veg
0.0264
0.0152
0.8538
0.1043
0.0003
0.0000
tree
0.0197
0.0056
0.1325
0.8422
0.0000
0.0000
car
0.2619
0.0004
0.0018
0.0072
0.7287
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.860
0.943
0.790
0.885
0.689
--
Recall/Completeness
0.824
0.945
0.854
0.842
0.729
--
F1
0.842
0.944
0.821
0.863
0.708
--

Overall accuracy 0.861

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
0.8593
0.0185
0.0903
0.0304
0.0014
0.0000
building
0.0250
0.9642
0.0061
0.0039
0.0008
0.0000
low_veg
0.0226
0.0091
0.8787
0.0894
0.0002
0.0000
tree
0.0128
0.0016
0.1174
0.8682
0.0000
0.0000
car
0.1974
0.0000
0.0000
0.0008
0.8017
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.889
0.966
0.818
0.909
0.745
--
Recall/Completeness
0.859
0.964
0.879
0.868
0.802
--
F1
0.874
0.965
0.847
0.888
0.772
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8716
0.0273
0.0597
0.0373
0.0041
0.0000
building
0.0431
0.9227
0.0297
0.0045
0.0000
0.0000
low_veg
0.0400
0.0276
0.7608
0.1716
0.0000
0.0000
tree
0.0071
0.0024
0.0565
0.9340
0.0000
0.0000
car
0.2483
0.0384
0.0112
0.0082
0.6939
0.0000
clutter
0.5359
0.4629
0.0010
0.0002
0.0000
0.0000
Precision/Correctness
0.841
0.860
0.786
0.902
0.777
-nan
Recall/Completeness
0.872
0.923
0.761
0.934
0.694
0.000
F1
0.856
0.890
0.773
0.918
0.733
-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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9040
0.0194
0.0545
0.0198
0.0023
0.0000
building
0.0412
0.9336
0.0239
0.0013
0.0001
0.0000
low_veg
0.0321
0.0176
0.8060
0.1443
0.0000
0.0000
tree
0.0027
0.0006
0.0440
0.9527
0.0000
0.0000
car
0.1787
0.0412
0.0025
0.0022
0.7754
0.0000
clutter
0.5412
0.4586
0.0003
0.0000
0.0000
0.0000
Precision/Correctness
0.859
0.884
0.817
0.928
0.833
-nan
Recall/Completeness
0.904
0.934
0.806
0.953
0.775
0.000
F1
0.881
0.908
0.811
0.940
0.803
-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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8868
0.0263
0.0529
0.0274
0.0066
0.0000
building
0.0457
0.9392
0.0095
0.0052
0.0004
0.0000
low_veg
0.0835
0.0324
0.7466
0.1374
0.0001
0.0000
tree
0.0193
0.0046
0.0807
0.8954
0.0000
0.0000
car
0.2854
0.0116
0.0116
0.0060
0.6854
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.894
0.911
0.782
0.873
0.772
--
Recall/Completeness
0.887
0.939
0.747
0.895
0.685
--
F1
0.890
0.925
0.764
0.884
0.726
--

Overall accuracy 0.868

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9185
0.0200
0.0441
0.0133
0.0041
0.0000
building
0.0402
0.9539
0.0028
0.0028
0.0004
0.0000
low_veg
0.0796
0.0206
0.7889
0.1109
0.0000
0.0000
tree
0.0070
0.0013
0.0618
0.9299
0.0000
0.0000
car
0.1963
0.0121
0.0107
0.0014
0.7795
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.918
0.940
0.825
0.910
0.820
--
Recall/Completeness
0.918
0.954
0.789
0.930
0.779
--
F1
0.918
0.947
0.807
0.920
0.799
--

Overall accuracy 0.902

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8278
0.0337
0.0991
0.0315
0.0078
0.0000
building
0.0788
0.8747
0.0311
0.0154
0.0001
0.0000
low_veg
0.0528
0.0197
0.8654
0.0620
0.0001
0.0000
tree
0.0183
0.0071
0.1838
0.7907
0.0000
0.0000
car
0.2310
0.0000
0.0005
0.0040
0.7644
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.741
0.917
0.842
0.859
0.707
--
Recall/Completeness
0.828
0.875
0.865
0.791
0.764
--
F1
0.782
0.896
0.853
0.823
0.735
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8601
0.0234
0.0892
0.0216
0.0058
0.0000
building
0.0757
0.8870
0.0238
0.0136
0.0000
0.0000
low_veg
0.0500
0.0157
0.8816
0.0525
0.0001
0.0000
tree
0.0121
0.0033
0.1611
0.8235
0.0000
0.0000
car
0.1401
0.0000
0.0000
0.0001
0.8599
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.761
0.941
0.865
0.881
0.725
--
Recall/Completeness
0.860
0.887
0.882
0.823
0.860
--
F1
0.807
0.913
0.873
0.852
0.787
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9447
0.0209
0.0233
0.0060
0.0052
0.0000
building
0.0838
0.8935
0.0175
0.0044
0.0008
0.0000
low_veg
0.0818
0.0212
0.7138
0.1831
0.0001
0.0000
tree
0.0116
0.0051
0.0603
0.9230
0.0000
0.0000
car
0.4292
0.0145
0.0075
0.0167
0.5320
0.0000
clutter
0.7938
0.0505
0.0666
0.0008
0.0883
0.0000
Precision/Correctness
0.829
0.963
0.875
0.699
0.888
-nan
Recall/Completeness
0.945
0.894
0.714
0.923
0.532
0.000
F1
0.883
0.927
0.786
0.796
0.665
-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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9676
0.0125
0.0162
0.0020
0.0017
0.0000
building
0.0713
0.9146
0.0112
0.0022
0.0007
0.0000
low_veg
0.0632
0.0092
0.7549
0.1727
0.0000
0.0000
tree
0.0047
0.0017
0.0435
0.9501
0.0000
0.0000
car
0.3305
0.0152
0.0061
0.0180
0.6303
0.0000
clutter
0.8085
0.0505
0.0442
0.0000
0.0968
0.0000
Precision/Correctness
0.865
0.981
0.912
0.729
0.939
-nan
Recall/Completeness
0.968
0.915
0.755
0.950
0.630
0.000
F1
0.913
0.947
0.826
0.825
0.754
-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
0.9415
0.0171
0.0264
0.0111
0.0040
0.0000
building
0.0783
0.9040
0.0113
0.0063
0.0001
0.0000
low_veg
0.1011
0.0214
0.6139
0.2632
0.0003
0.0000
tree
0.0054
0.0023
0.0362
0.9561
0.0000
0.0000
car
0.3324
0.0024
0.0073
0.0098
0.6481
0.0000
clutter
0.5373
0.3089
0.0818
0.0148
0.0572
0.0000
Precision/Correctness
0.875
0.925
0.814
0.695
0.716
-nan
Recall/Completeness
0.941
0.904
0.614
0.956
0.648
0.000
F1
0.907
0.914
0.700
0.805
0.680
-nan

Overall accuracy 0.852

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9622
0.0114
0.0188
0.0048
0.0028
0.0000
building
0.0724
0.9180
0.0062
0.0034
0.0000
0.0000
low_veg
0.0834
0.0093
0.6525
0.2547
0.0002
0.0000
tree
0.0016
0.0006
0.0275
0.9702
0.0000
0.0000
car
0.2315
0.0004
0.0075
0.0071
0.7535
0.0000
clutter
0.5252
0.3262
0.0798
0.0065
0.0623
0.0000
Precision/Correctness
0.895
0.945
0.856
0.730
0.724
-nan
Recall/Completeness
0.962
0.918
0.652
0.970
0.754
0.000
F1
0.928
0.931
0.741
0.833
0.739
-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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9607
0.0162
0.0173
0.0031
0.0027
0.0000
building
0.2580
0.7293
0.0096
0.0026
0.0005
0.0000
low_veg
0.0391
0.0031
0.9381
0.0197
0.0000
0.0000
tree
0.0379
0.0039
0.1067
0.8514
0.0000
0.0000
car
0.2422
0.2780
0.0127
0.0100
0.4572
0.0000
clutter
0.5545
0.3778
0.0348
0.0162
0.0166
0.0000
Precision/Correctness
0.774
0.871
0.858
0.944
0.903
-nan
Recall/Completeness
0.961
0.729
0.938
0.851
0.457
0.000
F1
0.857
0.794
0.896
0.895
0.607
-nan

Overall accuracy 0.830

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
0.9747
0.0109
0.0129
0.0004
0.0010
0.0000
building
0.2527
0.7424
0.0040
0.0005
0.0004
0.0000
low_veg
0.0239
0.0012
0.9608
0.0141
0.0000
0.0000
tree
0.0131
0.0007
0.0852
0.9011
0.0000
0.0000
car
0.1540
0.3077
0.0078
0.0090
0.5215
0.0000
clutter
0.5099
0.4283
0.0283
0.0138
0.0198
0.0000
Precision/Correctness
0.795
0.891
0.896
0.970
0.931
-nan
Recall/Completeness
0.975
0.742
0.961
0.901
0.522
0.000
F1
0.876
0.810
0.927
0.934
0.668
-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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9762
0.0093
0.0111
0.0015
0.0019
0.0000
building
0.2470
0.6556
0.0753
0.0198
0.0023
0.0000
low_veg
0.0996
0.0042
0.8661
0.0289
0.0012
0.0000
tree
0.0477
0.0028
0.1748
0.7747
0.0000
0.0000
car
0.4101
0.0018
0.0052
0.0185
0.5644
0.0000
clutter
0.8122
0.1441
0.0055
0.0045
0.0337
0.0000
Precision/Correctness
0.727
0.939
0.749
0.878
0.680
-nan
Recall/Completeness
0.976
0.656
0.866
0.775
0.564
0.000
F1
0.833
0.772
0.803
0.823
0.617
-nan

Overall accuracy 0.787

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
0.9866
0.0063
0.0060
0.0002
0.0009
0.0000
building
0.2429
0.6630
0.0730
0.0188
0.0023
0.0000
low_veg
0.0681
0.0011
0.9018
0.0278
0.0012
0.0000
tree
0.0211
0.0008
0.1573
0.8208
0.0000
0.0000
car
0.3286
0.0008
0.0037
0.0227
0.6443
0.0000
clutter
0.8010
0.1528
0.0045
0.0048
0.0369
0.0000
Precision/Correctness
0.747
0.951
0.771
0.890
0.676
-nan
Recall/Completeness
0.987
0.663
0.902
0.821
0.644
0.000
F1
0.850
0.781
0.831
0.854
0.660
-nan

Overall accuracy 0.807

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
0.8661
0.0298
0.0917
0.0107
0.0017
0.0000
building
0.0430
0.9322
0.0212
0.0032
0.0004
0.0000
low_veg
0.0399
0.0111
0.7646
0.1844
0.0000
0.0000
tree
0.0273
0.0063
0.1557
0.8107
0.0000
0.0000
car
0.3112
0.0013
0.0003
0.0046
0.6825
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.878
0.955
0.744
0.794
0.906
--
Recall/Completeness
0.866
0.932
0.765
0.811
0.683
--
F1
0.872
0.943
0.754
0.802
0.779
--

Overall accuracy 0.844

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
0.9117
0.0247
0.0608
0.0024
0.0003
0.0000
building
0.0323
0.9574
0.0092
0.0007
0.0004
0.0000
low_veg
0.0259
0.0044
0.7884
0.1813
0.0000
0.0000
tree
0.0089
0.0011
0.1201
0.8699
0.0000
0.0000
car
0.2125
0.0007
0.0000
0.0008
0.7860
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.923
0.973
0.811
0.815
0.958
--
Recall/Completeness
0.912
0.957
0.788
0.870
0.786
--
F1
0.917
0.965
0.800
0.841
0.863
--

Overall accuracy 0.884

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9028
0.0280
0.0562
0.0096
0.0035
0.0000
building
0.0374
0.9182
0.0396
0.0042
0.0006
0.0000
low_veg
0.0541
0.0122
0.8728
0.0608
0.0001
0.0000
tree
0.0223
0.0039
0.1545
0.8193
0.0000
0.0000
car
0.3218
0.0035
0.0543
0.0027
0.6177
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.883
0.952
0.814
0.885
0.842
--
Recall/Completeness
0.903
0.918
0.873
0.819
0.618
--
F1
0.893
0.935
0.842
0.851
0.713
--

Overall accuracy 0.879

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
0.9299
0.0255
0.0400
0.0032
0.0015
0.0000
building
0.0287
0.9434
0.0256
0.0017
0.0006
0.0000
low_veg
0.0391
0.0075
0.9043
0.0490
0.0001
0.0000
tree
0.0088
0.0014
0.1244
0.8655
0.0000
0.0000
car
0.2343
0.0023
0.0525
0.0001
0.7108
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.920
0.965
0.858
0.917
0.898
--
Recall/Completeness
0.930
0.943
0.904
0.865
0.711
--
F1
0.925
0.954
0.880
0.891
0.793
--

Overall accuracy 0.913

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image