Detailled semantic labeling (2D) result


Name Y. Liu et al.
Affiliation Chinese Academy of Sciences, China
Abbreviation CASIA
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
0.916
0.029
0.038
0.015
0.002
0.000
building
0.041
0.936
0.019
0.003
0.000
0.000
low_veg
0.052
0.024
0.801
0.123
0.000
0.000
tree
0.021
0.007
0.103
0.870
0.000
0.000
car
0.351
0.064
0.013
0.005
0.568
0.000
clutter
0.403
0.313
0.029
0.006
0.007
0.242
Precision/Correctness
0.880
0.930
0.815
0.862
0.921
0.965
Recall/Completeness
0.916
0.936
0.801
0.870
0.568
0.242
F1
0.898
0.933
0.808
0.866
0.703
0.387

Overall accuracy 0.875

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
0.942
0.020
0.031
0.006
0.001
0.000
building
0.034
0.951
0.013
0.001
0.000
0.000
low_veg
0.040
0.013
0.838
0.110
0.000
0.000
tree
0.009
0.002
0.085
0.904
0.000
0.000
car
0.214
0.071
0.009
0.002
0.703
0.000
clutter
0.366
0.330
0.026
0.002
0.007
0.270
Precision/Correctness
0.914
0.954
0.849
0.889
0.949
0.974
Recall/Completeness
0.942
0.951
0.838
0.904
0.703
0.270
F1
0.927
0.953
0.843
0.896
0.808
0.422

Overall accuracy 0.906
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9534
0.0141
0.0200
0.0099
0.0026
0.0000
building
0.0373
0.9435
0.0166
0.0018
0.0000
0.0008
low_veg
0.0652
0.0242
0.7568
0.1538
0.0001
0.0000
tree
0.0290
0.0072
0.0955
0.8671
0.0001
0.0011
car
0.3796
0.0153
0.0013
0.0013
0.6024
0.0000
clutter
0.0032
0.0000
0.0000
0.0140
0.0000
0.9828
Precision/Correctness
0.864
0.950
0.851
0.840
0.932
0.976
Recall/Completeness
0.953
0.944
0.757
0.867
0.602
0.983
F1
0.906
0.947
0.801
0.853
0.732
0.979

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9741
0.0099
0.0114
0.0024
0.0022
0.0000
building
0.0289
0.9601
0.0099
0.0002
0.0000
0.0009
low_veg
0.0470
0.0128
0.7915
0.1486
0.0000
0.0000
tree
0.0126
0.0026
0.0809
0.9035
0.0000
0.0003
car
0.2137
0.0148
0.0000
0.0003
0.7712
0.0000
clutter
0.0033
0.0000
0.0000
0.0025
0.0000
0.9942
Precision/Correctness
0.909
0.973
0.882
0.861
0.943
0.986
Recall/Completeness
0.974
0.960
0.791
0.904
0.771
0.994
F1
0.941
0.966
0.834
0.882
0.849
0.990

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9420
0.0410
0.0127
0.0033
0.0009
0.0000
building
0.0629
0.9271
0.0086
0.0011
0.0002
0.0000
low_veg
0.1239
0.0501
0.5604
0.2633
0.0024
0.0000
tree
0.0485
0.0148
0.1054
0.8312
0.0000
0.0000
car
0.4587
0.0213
0.0020
0.0017
0.5163
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.847
0.972
0.646
0.749
0.944
--
Recall/Completeness
0.942
0.927
0.560
0.831
0.516
--
F1
0.892
0.949
0.600
0.788
0.667
--

Overall accuracy 0.904

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
0.9706
0.0188
0.0094
0.0010
0.0002
0.0000
building
0.0517
0.9402
0.0076
0.0004
0.0002
0.0000
low_veg
0.0865
0.0161
0.6040
0.2912
0.0023
0.0000
tree
0.0191
0.0040
0.0991
0.8778
0.0000
0.0000
car
0.3261
0.0156
0.0010
0.0003
0.6571
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.878
0.989
0.672
0.772
0.966
--
Recall/Completeness
0.971
0.940
0.604
0.878
0.657
--
F1
0.922
0.964
0.636
0.822
0.782
--

Overall accuracy 0.929

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
0.9045
0.0595
0.0273
0.0076
0.0010
0.0000
building
0.0368
0.9491
0.0116
0.0025
0.0001
0.0000
low_veg
0.0790
0.0401
0.7485
0.1321
0.0002
0.0000
tree
0.0162
0.0050
0.0748
0.9040
0.0000
0.0000
car
0.4636
0.0333
0.0021
0.0010
0.5000
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.875
0.936
0.711
0.939
0.963
--
Recall/Completeness
0.905
0.949
0.749
0.904
0.500
--
F1
0.890
0.943
0.730
0.921
0.658
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9393
0.0389
0.0193
0.0022
0.0004
0.0000
building
0.0299
0.9614
0.0080
0.0007
0.0000
0.0000
low_veg
0.0633
0.0209
0.8004
0.1154
0.0000
0.0000
tree
0.0044
0.0012
0.0618
0.9325
0.0000
0.0000
car
0.3253
0.0290
0.0012
0.0000
0.6444
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.915
0.964
0.750
0.958
0.988
--
Recall/Completeness
0.939
0.961
0.800
0.932
0.644
--
F1
0.927
0.963
0.775
0.945
0.780
--

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
0.8903
0.0331
0.0672
0.0083
0.0010
0.0002
building
0.0321
0.9546
0.0108
0.0024
0.0000
0.0000
low_veg
0.0438
0.0211
0.8694
0.0656
0.0001
0.0000
tree
0.0246
0.0091
0.0900
0.8763
0.0000
0.0000
car
0.3928
0.0295
0.0304
0.0039
0.5433
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.907
0.928
0.789
0.929
0.964
--
Recall/Completeness
0.890
0.955
0.869
0.876
0.543
--
F1
0.899
0.941
0.827
0.902
0.695
--

Overall accuracy 0.892

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
0.9167
0.0232
0.0572
0.0022
0.0004
0.0002
building
0.0237
0.9704
0.0049
0.0009
0.0000
0.0000
low_veg
0.0269
0.0113
0.9035
0.0582
0.0000
0.0000
tree
0.0088
0.0041
0.0686
0.9185
0.0000
0.0000
car
0.2650
0.0298
0.0182
0.0027
0.6843
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.946
0.955
0.831
0.947
0.984
--
Recall/Completeness
0.917
0.970
0.904
0.918
0.684
--
F1
0.931
0.963
0.866
0.933
0.807
--

Overall accuracy 0.924

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
0.8891
0.0478
0.0464
0.0155
0.0012
0.0000
building
0.0572
0.9143
0.0217
0.0067
0.0000
0.0001
low_veg
0.0732
0.0241
0.7746
0.1279
0.0002
0.0000
tree
0.0123
0.0087
0.0939
0.8851
0.0000
0.0000
car
0.3326
0.0362
0.0131
0.0015
0.6166
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.914
0.899
0.765
0.859
0.903
--
Recall/Completeness
0.889
0.914
0.775
0.885
0.617
--
F1
0.902
0.907
0.770
0.872
0.733
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9228
0.0336
0.0375
0.0056
0.0005
0.0000
building
0.0524
0.9255
0.0176
0.0044
0.0000
0.0002
low_veg
0.0686
0.0095
0.8192
0.1026
0.0001
0.0000
tree
0.0042
0.0038
0.0713
0.9206
0.0000
0.0000
car
0.2017
0.0428
0.0081
0.0000
0.7474
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.931
0.935
0.808
0.900
0.944
--
Recall/Completeness
0.923
0.926
0.819
0.921
0.747
--
F1
0.927
0.930
0.813
0.910
0.834
--

Overall accuracy 0.905

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
0.8894
0.0290
0.0562
0.0244
0.0010
0.0000
building
0.0606
0.9185
0.0172
0.0036
0.0000
0.0000
low_veg
0.0501
0.0190
0.8002
0.1307
0.0000
0.0000
tree
0.0094
0.0071
0.1261
0.8575
0.0000
0.0000
car
0.3787
0.0079
0.0002
0.0002
0.6129
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.801
0.935
0.800
0.883
0.955
--
Recall/Completeness
0.889
0.919
0.800
0.857
0.613
--
F1
0.843
0.926
0.800
0.870
0.747
--

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_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9153
0.0264
0.0471
0.0110
0.0002
0.0000
building
0.0586
0.9307
0.0095
0.0012
0.0000
0.0000
low_veg
0.0472
0.0083
0.8359
0.1086
0.0000
0.0000
tree
0.0036
0.0024
0.1085
0.8855
0.0000
0.0000
car
0.2144
0.0096
0.0000
0.0000
0.7761
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.829
0.963
0.829
0.910
0.989
--
Recall/Completeness
0.915
0.931
0.836
0.885
0.776
--
F1
0.870
0.946
0.832
0.898
0.870
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8745
0.0242
0.0682
0.0303
0.0029
0.0000
building
0.0283
0.9501
0.0177
0.0039
0.0000
0.0000
low_veg
0.0864
0.0482
0.7290
0.1354
0.0002
0.0008
tree
0.0233
0.0125
0.1325
0.8316
0.0000
0.0000
car
0.2761
0.0000
0.0014
0.0055
0.7169
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.850
0.884
0.746
0.869
0.726
--
Recall/Completeness
0.874
0.950
0.729
0.832
0.717
--
F1
0.862
0.916
0.737
0.850
0.721
--

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_area14_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8950
0.0189
0.0635
0.0200
0.0026
0.0000
building
0.0238
0.9655
0.0098
0.0009
0.0000
0.0000
low_veg
0.0807
0.0324
0.7613
0.1245
0.0002
0.0009
tree
0.0145
0.0058
0.1193
0.8604
0.0000
0.0000
car
0.1331
0.0000
0.0002
0.0007
0.8660
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.877
0.921
0.769
0.890
0.724
--
Recall/Completeness
0.895
0.965
0.761
0.860
0.866
--
F1
0.886
0.943
0.765
0.875
0.788
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8463
0.0212
0.0820
0.0491
0.0014
0.0000
building
0.0149
0.9569
0.0199
0.0082
0.0000
0.0000
low_veg
0.0202
0.0171
0.8362
0.1265
0.0000
0.0000
tree
0.0216
0.0068
0.1131
0.8584
0.0000
0.0000
car
0.2575
0.0037
0.0118
0.0021
0.7249
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.884
0.940
0.811
0.871
0.881
--
Recall/Completeness
0.846
0.957
0.836
0.858
0.725
--
F1
0.865
0.948
0.824
0.865
0.795
--

Overall accuracy 0.867

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8812
0.0155
0.0730
0.0295
0.0007
0.0000
building
0.0109
0.9762
0.0101
0.0028
0.0000
0.0000
low_veg
0.0164
0.0093
0.8617
0.1127
0.0000
0.0000
tree
0.0140
0.0026
0.0991
0.8843
0.0000
0.0000
car
0.1465
0.0001
0.0052
0.0000
0.8482
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.916
0.967
0.839
0.894
0.926
--
Recall/Completeness
0.881
0.976
0.862
0.884
0.848
--
F1
0.898
0.971
0.850
0.889
0.885
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8530
0.0447
0.0661
0.0353
0.0010
0.0000
building
0.0045
0.9770
0.0149
0.0036
0.0000
0.0000
low_veg
0.0523
0.0322
0.7811
0.1341
0.0000
0.0003
tree
0.0082
0.0031
0.0556
0.9332
0.0000
0.0000
car
0.2330
0.0768
0.0459
0.0152
0.6291
0.0000
clutter
0.0000
1.0000
0.0000
0.0000
0.0000
0.0000
Precision/Correctness
0.919
0.785
0.796
0.919
0.931
0.000
Recall/Completeness
0.853
0.977
0.781
0.933
0.629
0.000
F1
0.885
0.870
0.788
0.926
0.751
-nan

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8873
0.0314
0.0627
0.0186
0.0001
0.0000
building
0.0030
0.9840
0.0123
0.0007
0.0000
0.0000
low_veg
0.0471
0.0152
0.8295
0.1078
0.0000
0.0003
tree
0.0034
0.0007
0.0425
0.9534
0.0000
0.0000
car
0.0999
0.0827
0.0402
0.0080
0.7692
0.0000
clutter
0.0000
1.0000
0.0000
0.0000
0.0000
0.0000
Precision/Correctness
0.942
0.812
0.825
0.944
0.995
0.000
Recall/Completeness
0.887
0.984
0.829
0.953
0.769
0.000
F1
0.914
0.890
0.827
0.948
0.868
-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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9086
0.0231
0.0376
0.0290
0.0017
0.0000
building
0.0271
0.9598
0.0103
0.0028
0.0000
0.0000
low_veg
0.0991
0.0356
0.7511
0.1141
0.0001
0.0000
tree
0.0227
0.0057
0.0605
0.9112
0.0000
0.0000
car
0.4239
0.0133
0.0184
0.0107
0.5336
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.890
0.913
0.828
0.889
0.909
--
Recall/Completeness
0.909
0.960
0.751
0.911
0.534
--
F1
0.899
0.936
0.788
0.900
0.672
--

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_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9398
0.0143
0.0309
0.0140
0.0011
0.0000
building
0.0222
0.9716
0.0056
0.0007
0.0000
0.0000
low_veg
0.0966
0.0177
0.7996
0.0859
0.0001
0.0000
tree
0.0092
0.0018
0.0435
0.9455
0.0000
0.0000
car
0.2941
0.0167
0.0107
0.0043
0.6742
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.916
0.953
0.869
0.927
0.937
--
Recall/Completeness
0.940
0.972
0.800
0.946
0.674
--
F1
0.928
0.962
0.833
0.936
0.784
--

Overall accuracy 0.918

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
0.8266
0.0374
0.1035
0.0298
0.0026
0.0000
building
0.0572
0.8870
0.0454
0.0104
0.0000
0.0000
low_veg
0.0232
0.0225
0.8863
0.0676
0.0001
0.0002
tree
0.0152
0.0077
0.1825
0.7946
0.0000
0.0000
car
0.3064
0.0000
0.0022
0.0014
0.6900
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.826
0.909
0.839
0.855
0.856
--
Recall/Completeness
0.827
0.887
0.886
0.795
0.690
--
F1
0.826
0.898
0.862
0.824
0.764
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8567
0.0261
0.0960
0.0190
0.0021
0.0000
building
0.0544
0.8976
0.0398
0.0082
0.0000
0.0000
low_veg
0.0202
0.0172
0.9035
0.0588
0.0001
0.0002
tree
0.0084
0.0037
0.1596
0.8284
0.0000
0.0000
car
0.1479
0.0000
0.0001
0.0000
0.8520
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.852
0.935
0.861
0.878
0.866
--
Recall/Completeness
0.857
0.898
0.903
0.828
0.852
--
F1
0.854
0.916
0.881
0.852
0.859
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9333
0.0370
0.0199
0.0055
0.0042
0.0000
building
0.0420
0.9410
0.0147
0.0023
0.0000
0.0000
low_veg
0.0668
0.0408
0.6990
0.1931
0.0003
0.0000
tree
0.0110
0.0071
0.0657
0.9161
0.0001
0.0000
car
0.3242
0.0152
0.0071
0.0054
0.6482
0.0000
clutter
0.3284
0.4656
0.1395
0.0048
0.0616
0.0000
Precision/Correctness
0.881
0.937
0.880
0.694
0.930
-nan
Recall/Completeness
0.933
0.941
0.699
0.916
0.648
0.000
F1
0.906
0.939
0.779
0.790
0.764
-nan

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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9605
0.0233
0.0134
0.0018
0.0011
0.0000
building
0.0335
0.9559
0.0101
0.0006
0.0000
0.0000
low_veg
0.0498
0.0218
0.7443
0.1840
0.0001
0.0000
tree
0.0038
0.0017
0.0472
0.9472
0.0000
0.0000
car
0.1960
0.0129
0.0044
0.0040
0.7827
0.0000
clutter
0.2871
0.5093
0.1262
0.0027
0.0747
0.0000
Precision/Correctness
0.918
0.964
0.916
0.722
0.974
-nan
Recall/Completeness
0.961
0.956
0.744
0.947
0.783
0.000
F1
0.939
0.960
0.821
0.819
0.868
-nan

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9455
0.0197
0.0202
0.0122
0.0023
0.0000
building
0.0533
0.9322
0.0087
0.0058
0.0000
0.0000
low_veg
0.0473
0.0235
0.6528
0.2760
0.0003
0.0000
tree
0.0048
0.0027
0.0303
0.9622
0.0000
0.0000
car
0.3820
0.0051
0.0137
0.0080
0.5898
0.0014
clutter
0.5138
0.4225
0.0439
0.0067
0.0087
0.0044
Precision/Correctness
0.904
0.911
0.860
0.687
0.860
0.873
Recall/Completeness
0.946
0.932
0.653
0.962
0.590
0.004
F1
0.925
0.921
0.742
0.802
0.700
0.009

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
0.9662
0.0127
0.0140
0.0051
0.0019
0.0000
building
0.0487
0.9440
0.0045
0.0028
0.0000
0.0000
low_veg
0.0269
0.0087
0.6983
0.2660
0.0002
0.0000
tree
0.0009
0.0005
0.0218
0.9767
0.0000
0.0000
car
0.2492
0.0010
0.0137
0.0045
0.7296
0.0020
clutter
0.5021
0.4465
0.0338
0.0034
0.0092
0.0051
Precision/Correctness
0.926
0.934
0.899
0.724
0.870
0.870
Recall/Completeness
0.966
0.944
0.698
0.977
0.730
0.005
F1
0.945
0.939
0.786
0.832
0.794
0.010

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9761
0.0089
0.0096
0.0037
0.0017
0.0000
building
0.0329
0.9564
0.0088
0.0016
0.0004
0.0000
low_veg
0.0375
0.0038
0.9254
0.0332
0.0001
0.0000
tree
0.0401
0.0048
0.0812
0.8735
0.0004
0.0000
car
0.2571
0.2798
0.0084
0.0056
0.4491
0.0001
clutter
0.6002
0.2370
0.0794
0.0127
0.0044
0.0662
Precision/Correctness
0.886
0.921
0.883
0.930
0.944
0.999
Recall/Completeness
0.976
0.956
0.925
0.874
0.449
0.066
F1
0.929
0.938
0.903
0.901
0.609
0.124

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9865
0.0046
0.0073
0.0006
0.0009
0.0000
building
0.0218
0.9727
0.0051
0.0002
0.0002
0.0000
low_veg
0.0192
0.0009
0.9533
0.0266
0.0000
0.0000
tree
0.0141
0.0007
0.0650
0.9202
0.0000
0.0000
car
0.1361
0.3135
0.0031
0.0021
0.5452
0.0000
clutter
0.5562
0.2667
0.0856
0.0066
0.0053
0.0796
Precision/Correctness
0.924
0.937
0.910
0.957
0.965
1.000
Recall/Completeness
0.987
0.973
0.953
0.920
0.545
0.080
F1
0.954
0.954
0.931
0.938
0.697
0.147

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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9744
0.0145
0.0079
0.0027
0.0003
0.0002
building
0.0489
0.9169
0.0312
0.0025
0.0002
0.0004
low_veg
0.0526
0.0216
0.8765
0.0490
0.0002
0.0000
tree
0.0430
0.0063
0.1374
0.8132
0.0000
0.0000
car
0.3831
0.0717
0.0030
0.0052
0.5369
0.0000
clutter
0.6228
0.1001
0.0033
0.0001
0.0102
0.2635
Precision/Correctness
0.852
0.944
0.835
0.901
0.899
0.988
Recall/Completeness
0.974
0.917
0.877
0.813
0.537
0.264
F1
0.909
0.930
0.855
0.855
0.672
0.416

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_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9866
0.0090
0.0037
0.0006
0.0001
0.0000
building
0.0405
0.9296
0.0287
0.0008
0.0002
0.0003
low_veg
0.0227
0.0118
0.9189
0.0464
0.0002
0.0000
tree
0.0164
0.0017
0.1257
0.8562
0.0000
0.0000
car
0.2235
0.0861
0.0007
0.0015
0.6882
0.0000
clutter
0.5917
0.0969
0.0018
0.0000
0.0109
0.2987
Precision/Correctness
0.886
0.962
0.858
0.921
0.913
0.995
Recall/Completeness
0.987
0.930
0.919
0.856
0.688
0.299
F1
0.933
0.945
0.887
0.887
0.785
0.460

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8838
0.0372
0.0613
0.0173
0.0003
0.0000
building
0.0315
0.9492
0.0158
0.0032
0.0000
0.0004
low_veg
0.0432
0.0131
0.7439
0.1998
0.0000
0.0000
tree
0.0306
0.0084
0.1081
0.8527
0.0000
0.0001
car
0.3968
0.0174
0.0000
0.0103
0.5755
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.883
0.945
0.804
0.785
0.981
--
Recall/Completeness
0.884
0.949
0.744
0.853
0.576
--
F1
0.884
0.947
0.773
0.817
0.725
--

Overall accuracy 0.857

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
0.9227
0.0299
0.0428
0.0046
0.0000
0.0000
building
0.0226
0.9696
0.0072
0.0003
0.0000
0.0004
low_veg
0.0278
0.0044
0.7765
0.1913
0.0000
0.0000
tree
0.0085
0.0021
0.0836
0.9057
0.0000
0.0001
car
0.2487
0.0126
0.0000
0.0039
0.7348
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.932
0.968
0.857
0.811
0.998
--
Recall/Completeness
0.923
0.970
0.776
0.906
0.735
--
F1
0.927
0.969
0.815
0.856
0.847
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9112
0.0250
0.0489
0.0133
0.0014
0.0001
building
0.0303
0.9165
0.0486
0.0034
0.0013
0.0000
low_veg
0.0389
0.0117
0.8772
0.0720
0.0002
0.0000
tree
0.0269
0.0040
0.1340
0.8350
0.0001
0.0000
car
0.2976
0.0264
0.0533
0.0070
0.6157
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.902
0.955
0.825
0.869
0.884
--
Recall/Completeness
0.911
0.916
0.877
0.835
0.616
--
F1
0.906
0.935
0.850
0.851
0.726
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9363
0.0223
0.0360
0.0049
0.0004
0.0001
building
0.0237
0.9409
0.0332
0.0011
0.0011
0.0000
low_veg
0.0215
0.0060
0.9146
0.0579
0.0000
0.0000
tree
0.0103
0.0011
0.1065
0.8821
0.0000
0.0000
car
0.1605
0.0278
0.0493
0.0025
0.7599
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.943
0.969
0.866
0.905
0.928
--
Recall/Completeness
0.936
0.941
0.915
0.882
0.760
--
F1
0.939
0.955
0.890
0.893
0.835
--

Overall accuracy 0.920

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image