Detailled semantic labeling (2D) result


Name C. Lyu and J. Jiang
Affiliation Beihang University, China
Abbreviation BUAA_2
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
0.913
0.028
0.041
0.015
0.004
0.000
building
0.053
0.923
0.020
0.004
0.000
0.000
low_veg
0.054
0.024
0.805
0.117
0.000
0.000
tree
0.023
0.007
0.110
0.860
0.000
0.000
car
0.248
0.029
0.007
0.005
0.711
0.000
clutter
0.689
0.261
0.016
0.011
0.022
0.000
Precision/Correctness
0.864
0.933
0.807
0.865
0.874
-nan
Recall/Completeness
0.913
0.923
0.805
0.860
0.711
0.000
F1
0.888
0.928
0.806
0.862
0.784
-nan

Overall accuracy 0.870

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
0.937
0.021
0.033
0.007
0.002
0.000
building
0.044
0.941
0.013
0.002
0.000
0.000
low_veg
0.043
0.015
0.839
0.103
0.000
0.000
tree
0.011
0.003
0.091
0.895
0.000
0.000
car
0.140
0.034
0.004
0.002
0.820
0.000
clutter
0.677
0.281
0.014
0.006
0.023
0.000
Precision/Correctness
0.895
0.953
0.842
0.892
0.905
-nan
Recall/Completeness
0.937
0.941
0.839
0.895
0.820
0.000
F1
0.916
0.947
0.840
0.893
0.860
-nan

Overall accuracy 0.899
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9567
0.0093
0.0185
0.0111
0.0045
0.0000
building
0.0485
0.9264
0.0219
0.0032
0.0000
0.0000
low_veg
0.0715
0.0193
0.7614
0.1478
0.0000
0.0000
tree
0.0286
0.0032
0.1074
0.8607
0.0001
0.0000
car
0.2639
0.0008
0.0002
0.0015
0.7337
0.0000
clutter
0.9658
0.0000
0.0125
0.0217
0.0000
0.0000
Precision/Correctness
0.806
0.964
0.837
0.841
0.908
-nan
Recall/Completeness
0.957
0.926
0.761
0.861
0.734
0.000
F1
0.875
0.945
0.797
0.851
0.812
-nan

Overall accuracy 0.858

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
0.9783
0.0060
0.0087
0.0033
0.0036
0.0000
building
0.0404
0.9477
0.0111
0.0007
0.0000
0.0000
low_veg
0.0532
0.0123
0.7936
0.1408
0.0000
0.0000
tree
0.0130
0.0010
0.0927
0.8933
0.0000
0.0000
car
0.1220
0.0001
0.0000
0.0006
0.8773
0.0000
clutter
0.9820
0.0000
0.0130
0.0050
0.0000
0.0000
Precision/Correctness
0.839
0.980
0.871
0.865
0.921
-nan
Recall/Completeness
0.978
0.948
0.794
0.893
0.877
0.000
F1
0.903
0.963
0.830
0.879
0.899
-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
0.9376
0.0416
0.0135
0.0041
0.0032
0.0000
building
0.0735
0.9148
0.0100
0.0016
0.0001
0.0000
low_veg
0.1193
0.0505
0.5598
0.2698
0.0005
0.0000
tree
0.0523
0.0180
0.1018
0.8279
0.0001
0.0000
car
0.3080
0.0119
0.0016
0.0013
0.6771
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.836
0.971
0.633
0.738
0.908
--
Recall/Completeness
0.938
0.915
0.560
0.828
0.677
--
F1
0.884
0.942
0.594
0.780
0.776
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9635
0.0234
0.0100
0.0016
0.0015
0.0000
building
0.0602
0.9303
0.0087
0.0008
0.0000
0.0000
low_veg
0.0828
0.0178
0.5993
0.3001
0.0000
0.0000
tree
0.0222
0.0067
0.0953
0.8757
0.0000
0.0000
car
0.1913
0.0071
0.0000
0.0007
0.8009
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.868
0.986
0.658
0.761
0.952
--
Recall/Completeness
0.963
0.930
0.599
0.876
0.801
--
F1
0.913
0.957
0.627
0.814
0.870
--

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_area6_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9191
0.0395
0.0303
0.0083
0.0028
0.0000
building
0.0483
0.9318
0.0177
0.0021
0.0001
0.0000
low_veg
0.0936
0.0275
0.7542
0.1244
0.0003
0.0000
tree
0.0209
0.0045
0.0735
0.9009
0.0002
0.0000
car
0.2812
0.0114
0.0003
0.0013
0.7058
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.865
0.955
0.698
0.942
0.931
--
Recall/Completeness
0.919
0.932
0.754
0.901
0.706
--
F1
0.891
0.943
0.725
0.921
0.803
--

Overall accuracy 0.897

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9530
0.0224
0.0216
0.0025
0.0004
0.0000
building
0.0391
0.9473
0.0131
0.0005
0.0000
0.0000
low_veg
0.0800
0.0114
0.7997
0.1090
0.0001
0.0000
tree
0.0106
0.0011
0.0587
0.9295
0.0001
0.0000
car
0.1721
0.0055
0.0000
0.0002
0.8221
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.900
0.979
0.740
0.960
0.987
--
Recall/Completeness
0.953
0.947
0.800
0.930
0.822
--
F1
0.926
0.963
0.769
0.945
0.897
--

Overall accuracy 0.928

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8876
0.0206
0.0792
0.0091
0.0035
0.0000
building
0.0347
0.9433
0.0185
0.0035
0.0000
0.0000
low_veg
0.0393
0.0158
0.8870
0.0578
0.0001
0.0000
tree
0.0234
0.0080
0.1126
0.8560
0.0000
0.0000
car
0.2736
0.0043
0.0319
0.0090
0.6812
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.915
0.950
0.755
0.931
0.912
--
Recall/Completeness
0.888
0.943
0.887
0.856
0.681
--
F1
0.901
0.946
0.816
0.892
0.780
--

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
0.9137
0.0137
0.0679
0.0026
0.0020
0.0000
building
0.0260
0.9618
0.0109
0.0013
0.0000
0.0000
low_veg
0.0245
0.0074
0.9189
0.0492
0.0000
0.0000
tree
0.0086
0.0038
0.0906
0.8971
0.0000
0.0000
car
0.1646
0.0018
0.0234
0.0077
0.8024
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.950
0.972
0.795
0.952
0.942
--
Recall/Completeness
0.914
0.962
0.919
0.897
0.802
--
F1
0.931
0.967
0.853
0.924
0.867
--

Overall accuracy 0.921

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
0.8392
0.0716
0.0692
0.0170
0.0030
0.0000
building
0.0479
0.9260
0.0171
0.0090
0.0000
0.0000
low_veg
0.0789
0.0339
0.7641
0.1230
0.0000
0.0000
tree
0.0155
0.0090
0.1010
0.8745
0.0000
0.0000
car
0.2240
0.0607
0.0043
0.0034
0.7076
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.913
0.862
0.726
0.856
0.815
--
Recall/Completeness
0.839
0.926
0.764
0.874
0.708
--
F1
0.875
0.893
0.744
0.865
0.757
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8675
0.0611
0.0623
0.0072
0.0019
0.0000
building
0.0430
0.9374
0.0129
0.0067
0.0000
0.0000
low_veg
0.0761
0.0213
0.8043
0.0983
0.0000
0.0000
tree
0.0062
0.0043
0.0778
0.9117
0.0000
0.0000
car
0.1322
0.0733
0.0019
0.0005
0.7920
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.930
0.889
0.761
0.897
0.849
--
Recall/Completeness
0.868
0.937
0.804
0.912
0.792
--
F1
0.898
0.912
0.782
0.904
0.820
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8905
0.0378
0.0423
0.0273
0.0021
0.0000
building
0.0588
0.9277
0.0093
0.0040
0.0002
0.0000
low_veg
0.0560
0.0250
0.7822
0.1367
0.0001
0.0000
tree
0.0118
0.0090
0.1173
0.8619
0.0000
0.0000
car
0.1746
0.0000
0.0008
0.0007
0.8238
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.794
0.917
0.816
0.878
0.923
--
Recall/Completeness
0.891
0.928
0.782
0.862
0.824
--
F1
0.840
0.923
0.799
0.870
0.871
--

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_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9156
0.0351
0.0347
0.0143
0.0002
0.0000
building
0.0582
0.9378
0.0028
0.0011
0.0002
0.0000
low_veg
0.0523
0.0155
0.8164
0.1158
0.0000
0.0000
tree
0.0062
0.0042
0.0985
0.8911
0.0000
0.0000
car
0.0542
0.0000
0.0000
0.0000
0.9458
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.820
0.943
0.846
0.904
0.983
--
Recall/Completeness
0.916
0.938
0.816
0.891
0.946
--
F1
0.865
0.941
0.831
0.898
0.964
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8367
0.0366
0.0902
0.0302
0.0064
0.0000
building
0.0076
0.9747
0.0135
0.0042
0.0000
0.0000
low_veg
0.0743
0.0549
0.7498
0.1207
0.0003
0.0000
tree
0.0247
0.0166
0.1353
0.8234
0.0000
0.0000
car
0.1174
0.0000
0.0000
0.0026
0.8800
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.869
0.860
0.736
0.878
0.602
--
Recall/Completeness
0.837
0.975
0.750
0.823
0.880
--
F1
0.853
0.914
0.743
0.850
0.715
--

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_area14_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8578
0.0308
0.0859
0.0200
0.0056
0.0000
building
0.0039
0.9887
0.0060
0.0014
0.0000
0.0000
low_veg
0.0696
0.0395
0.7799
0.1108
0.0002
0.0000
tree
0.0167
0.0089
0.1227
0.8518
0.0000
0.0000
car
0.0279
0.0000
0.0000
0.0000
0.9721
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.894
0.896
0.756
0.899
0.584
--
Recall/Completeness
0.858
0.989
0.780
0.852
0.972
--
F1
0.876
0.940
0.768
0.875
0.730
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8434
0.0331
0.0751
0.0455
0.0029
0.0000
building
0.0474
0.9279
0.0168
0.0077
0.0002
0.0000
low_veg
0.0204
0.0201
0.8615
0.0979
0.0001
0.0000
tree
0.0256
0.0066
0.1247
0.8431
0.0000
0.0000
car
0.1278
0.0063
0.0114
0.0064
0.8481
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.844
0.924
0.809
0.892
0.787
--
Recall/Completeness
0.843
0.928
0.861
0.843
0.848
--
F1
0.844
0.926
0.834
0.867
0.816
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8754
0.0284
0.0669
0.0278
0.0014
0.0000
building
0.0446
0.9462
0.0060
0.0030
0.0002
0.0000
low_veg
0.0160
0.0132
0.8855
0.0852
0.0001
0.0000
tree
0.0174
0.0030
0.1090
0.8705
0.0000
0.0000
car
0.0620
0.0030
0.0070
0.0015
0.9264
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.873
0.948
0.837
0.914
0.842
--
Recall/Completeness
0.875
0.946
0.886
0.871
0.926
--
F1
0.874
0.947
0.860
0.892
0.882
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8803
0.0266
0.0490
0.0391
0.0050
0.0000
building
0.0056
0.9806
0.0098
0.0040
0.0000
0.0000
low_veg
0.0590
0.0336
0.7713
0.1355
0.0005
0.0000
tree
0.0086
0.0031
0.0558
0.9325
0.0000
0.0000
car
0.1269
0.0027
0.0147
0.0138
0.8419
0.0000
clutter
0.0014
0.9983
0.0000
0.0004
0.0000
0.0000
Precision/Correctness
0.915
0.800
0.813
0.917
0.759
-nan
Recall/Completeness
0.880
0.981
0.771
0.932
0.842
0.000
F1
0.897
0.881
0.791
0.924
0.798
-nan

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9125
0.0169
0.0449
0.0224
0.0033
0.0000
building
0.0031
0.9907
0.0050
0.0012
0.0000
0.0000
low_veg
0.0538
0.0226
0.8135
0.1096
0.0004
0.0000
tree
0.0034
0.0010
0.0423
0.9532
0.0000
0.0000
car
0.0541
0.0006
0.0053
0.0079
0.9322
0.0000
clutter
0.0000
1.0000
0.0000
0.0000
0.0000
0.0000
Precision/Correctness
0.938
0.820
0.845
0.941
0.788
-nan
Recall/Completeness
0.912
0.991
0.814
0.953
0.932
0.000
F1
0.925
0.897
0.829
0.947
0.854
-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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8983
0.0268
0.0401
0.0299
0.0049
0.0000
building
0.0265
0.9573
0.0113
0.0048
0.0001
0.0000
low_veg
0.0932
0.0386
0.7519
0.1161
0.0001
0.0000
tree
0.0227
0.0043
0.0756
0.8974
0.0001
0.0000
car
0.2822
0.0017
0.0047
0.0087
0.7027
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.896
0.907
0.805
0.884
0.824
--
Recall/Completeness
0.898
0.957
0.752
0.897
0.703
--
F1
0.897
0.931
0.778
0.891
0.758
--

Overall accuracy 0.877

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9285
0.0204
0.0328
0.0152
0.0032
0.0000
building
0.0205
0.9736
0.0044
0.0015
0.0001
0.0000
low_veg
0.0911
0.0256
0.7925
0.0907
0.0001
0.0000
tree
0.0090
0.0015
0.0567
0.9328
0.0000
0.0000
car
0.1777
0.0005
0.0030
0.0021
0.8166
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.922
0.936
0.848
0.921
0.861
--
Recall/Completeness
0.928
0.974
0.792
0.933
0.817
--
F1
0.925
0.955
0.819
0.927
0.838
--

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
0.8079
0.0519
0.1036
0.0303
0.0064
0.0000
building
0.0638
0.8897
0.0351
0.0114
0.0000
0.0000
low_veg
0.0230
0.0227
0.8926
0.0615
0.0002
0.0000
tree
0.0209
0.0075
0.1730
0.7987
0.0000
0.0000
car
0.1663
0.0075
0.0002
0.0022
0.8239
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.811
0.899
0.848
0.864
0.752
--
Recall/Completeness
0.808
0.890
0.893
0.799
0.824
--
F1
0.810
0.894
0.870
0.830
0.786
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8373
0.0424
0.0964
0.0194
0.0046
0.0000
building
0.0619
0.9000
0.0290
0.0091
0.0000
0.0000
low_veg
0.0195
0.0171
0.9114
0.0517
0.0002
0.0000
tree
0.0142
0.0037
0.1504
0.8317
0.0000
0.0000
car
0.0754
0.0095
0.0000
0.0005
0.9146
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.835
0.924
0.870
0.888
0.766
--
Recall/Completeness
0.837
0.900
0.911
0.832
0.915
--
F1
0.836
0.912
0.890
0.859
0.834
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9445
0.0214
0.0220
0.0070
0.0052
0.0000
building
0.0728
0.9013
0.0224
0.0031
0.0003
0.0000
low_veg
0.0654
0.0262
0.7227
0.1850
0.0006
0.0000
tree
0.0135
0.0074
0.0497
0.9292
0.0002
0.0000
car
0.2609
0.0077
0.0021
0.0056
0.7237
0.0000
clutter
0.5793
0.1975
0.1239
0.0182
0.0812
0.0000
Precision/Correctness
0.859
0.959
0.875
0.701
0.917
-nan
Recall/Completeness
0.944
0.901
0.723
0.929
0.724
0.000
F1
0.900
0.929
0.792
0.799
0.809
-nan

Overall accuracy 0.875

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
0.9681
0.0127
0.0153
0.0026
0.0013
0.0000
building
0.0616
0.9209
0.0162
0.0011
0.0002
0.0000
low_veg
0.0488
0.0132
0.7640
0.1735
0.0005
0.0000
tree
0.0056
0.0032
0.0334
0.9578
0.0000
0.0000
car
0.1478
0.0056
0.0007
0.0037
0.8422
0.0000
clutter
0.5752
0.2006
0.1163
0.0103
0.0976
0.0000
Precision/Correctness
0.893
0.979
0.910
0.733
0.963
-nan
Recall/Completeness
0.968
0.921
0.764
0.958
0.842
0.000
F1
0.929
0.949
0.831
0.830
0.899
-nan

Overall accuracy 0.906

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
0.9457
0.0134
0.0236
0.0128
0.0044
0.0000
building
0.0487
0.9339
0.0112
0.0060
0.0002
0.0000
low_veg
0.0555
0.0220
0.6370
0.2853
0.0001
0.0000
tree
0.0061
0.0023
0.0293
0.9623
0.0000
0.0000
car
0.2695
0.0023
0.0077
0.0070
0.7134
0.0000
clutter
0.5938
0.3467
0.0311
0.0159
0.0124
0.0000
Precision/Correctness
0.902
0.929
0.845
0.679
0.805
-nan
Recall/Completeness
0.946
0.934
0.637
0.962
0.713
0.000
F1
0.924
0.931
0.726
0.796
0.757
-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
0.9653
0.0079
0.0174
0.0059
0.0034
0.0000
building
0.0427
0.9481
0.0060
0.0029
0.0002
0.0000
low_veg
0.0333
0.0097
0.6832
0.2738
0.0000
0.0000
tree
0.0015
0.0007
0.0210
0.9768
0.0000
0.0000
car
0.1790
0.0002
0.0083
0.0040
0.8085
0.0000
clutter
0.5764
0.3754
0.0277
0.0072
0.0133
0.0000
Precision/Correctness
0.924
0.947
0.884
0.717
0.815
-nan
Recall/Completeness
0.965
0.948
0.683
0.977
0.809
0.000
F1
0.944
0.948
0.771
0.827
0.812
-nan

Overall accuracy 0.895

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9776
0.0058
0.0099
0.0031
0.0036
0.0000
building
0.0599
0.9281
0.0091
0.0014
0.0015
0.0000
low_veg
0.0397
0.0022
0.9394
0.0186
0.0001
0.0000
tree
0.0425
0.0049
0.1125
0.8395
0.0007
0.0000
car
0.2196
0.1236
0.0092
0.0047
0.6428
0.0000
clutter
0.7759
0.1737
0.0183
0.0142
0.0179
0.0000
Precision/Correctness
0.861
0.950
0.875
0.950
0.902
-nan
Recall/Completeness
0.978
0.928
0.939
0.839
0.643
0.000
F1
0.916
0.939
0.906
0.891
0.751
-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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9868
0.0043
0.0067
0.0006
0.0015
0.0000
building
0.0442
0.9513
0.0030
0.0002
0.0013
0.0000
low_veg
0.0212
0.0012
0.9654
0.0122
0.0000
0.0000
tree
0.0141
0.0016
0.0925
0.8917
0.0002
0.0000
car
0.1310
0.1473
0.0045
0.0010
0.7161
0.0000
clutter
0.7568
0.1996
0.0161
0.0084
0.0191
0.0000
Precision/Correctness
0.897
0.957
0.910
0.976
0.929
-nan
Recall/Completeness
0.987
0.951
0.965
0.892
0.716
0.000
F1
0.940
0.954
0.937
0.932
0.809
-nan

Overall accuracy 0.925

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
0.9830
0.0077
0.0059
0.0025
0.0009
0.0000
building
0.0641
0.8737
0.0568
0.0032
0.0022
0.0000
low_veg
0.0713
0.0089
0.8313
0.0885
0.0000
0.0000
tree
0.0638
0.0070
0.1380
0.7912
0.0000
0.0000
car
0.2583
0.0405
0.0112
0.0003
0.6897
0.0000
clutter
0.9050
0.0318
0.0061
0.0057
0.0513
0.0000
Precision/Correctness
0.811
0.972
0.792
0.839
0.702
-nan
Recall/Completeness
0.983
0.874
0.831
0.791
0.690
0.000
F1
0.888
0.920
0.811
0.815
0.696
-nan

Overall accuracy 0.851

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9912
0.0060
0.0020
0.0007
0.0002
0.0000
building
0.0531
0.8890
0.0547
0.0012
0.0020
0.0000
low_veg
0.0392
0.0032
0.8657
0.0920
0.0000
0.0000
tree
0.0370
0.0030
0.1254
0.8347
0.0000
0.0000
car
0.1332
0.0495
0.0079
0.0000
0.8094
0.0000
clutter
0.9011
0.0323
0.0048
0.0063
0.0555
0.0000
Precision/Correctness
0.841
0.981
0.812
0.852
0.699
-nan
Recall/Completeness
0.991
0.889
0.866
0.835
0.809
0.000
F1
0.910
0.933
0.838
0.843
0.750
-nan

Overall accuracy 0.874

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
0.8776
0.0371
0.0687
0.0161
0.0005
0.0000
building
0.0621
0.9178
0.0162
0.0039
0.0000
0.0000
low_veg
0.0479
0.0138
0.7529
0.1854
0.0000
0.0000
tree
0.0372
0.0071
0.1390
0.8166
0.0001
0.0000
car
0.2878
0.0007
0.0008
0.0080
0.7027
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.849
0.944
0.775
0.789
0.973
--
Recall/Completeness
0.878
0.918
0.753
0.817
0.703
--
F1
0.863
0.931
0.764
0.803
0.816
--

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_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9124
0.0341
0.0488
0.0048
0.0000
0.0000
building
0.0493
0.9435
0.0064
0.0008
0.0000
0.0000
low_veg
0.0331
0.0059
0.7812
0.1798
0.0000
0.0000
tree
0.0145
0.0022
0.1103
0.8730
0.0000
0.0000
car
0.1541
0.0001
0.0000
0.0030
0.8429
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.896
0.963
0.831
0.814
0.999
--
Recall/Completeness
0.912
0.943
0.781
0.873
0.843
--
F1
0.904
0.953
0.805
0.843
0.914
--

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
0.9202
0.0250
0.0408
0.0116
0.0024
0.0000
building
0.0463
0.9162
0.0341
0.0032
0.0001
0.0000
low_veg
0.0426
0.0188
0.8798
0.0586
0.0002
0.0000
tree
0.0256
0.0053
0.1496
0.8194
0.0001
0.0000
car
0.1987
0.0291
0.0146
0.0060
0.7516
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.890
0.946
0.834
0.887
0.906
--
Recall/Completeness
0.920
0.916
0.880
0.819
0.752
--
F1
0.905
0.931
0.856
0.852
0.821
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9453
0.0222
0.0276
0.0040
0.0009
0.0000
building
0.0389
0.9388
0.0212
0.0010
0.0001
0.0000
low_veg
0.0281
0.0122
0.9143
0.0452
0.0001
0.0000
tree
0.0094
0.0016
0.1239
0.8651
0.0000
0.0000
car
0.0981
0.0330
0.0035
0.0020
0.8634
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.925
0.962
0.874
0.922
0.953
--
Recall/Completeness
0.945
0.939
0.914
0.865
0.863
--
F1
0.935
0.950
0.894
0.893
0.906
--

Overall accuracy 0.919

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image