Detailled semantic labeling (2D) result


Name M. Volpi and D. Tuia
Affiliation MMRS, Department of Geography, University of Zurich, Switzerland
Abbreviation UZ_1
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
0.884
0.046
0.044
0.021
0.004
0.000
building
0.064
0.908
0.013
0.013
0.001
0.000
low_veg
0.065
0.037
0.784
0.113
0.001
0.000
tree
0.015
0.010
0.143
0.833
0.000
0.000
car
0.542
0.045
0.019
0.007
0.387
0.001
clutter
0.558
0.254
0.057
0.052
0.057
0.021
Precision/Correctness
0.843
0.901
0.777
0.845
0.717
0.536
Recall/Completeness
0.884
0.908
0.784
0.833
0.387
0.021
F1
0.863
0.905
0.780
0.839
0.503
0.041

Overall accuracy 0.843

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
0.916
0.037
0.034
0.011
0.002
0.000
building
0.058
0.924
0.007
0.010
0.001
0.000
low_veg
0.055
0.026
0.820
0.098
0.001
0.000
tree
0.007
0.005
0.126
0.861
0.000
0.000
car
0.467
0.052
0.013
0.003
0.463
0.001
clutter
0.536
0.275
0.054
0.052
0.060
0.024
Precision/Correctness
0.870
0.926
0.812
0.878
0.750
0.560
Recall/Completeness
0.916
0.924
0.820
0.861
0.463
0.024
F1
0.892
0.925
0.816
0.869
0.573
0.046

Overall accuracy 0.873
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9168
0.0195
0.0428
0.0180
0.0028
0.0001
building
0.0232
0.9483
0.0219
0.0067
0.0000
0.0000
low_veg
0.0865
0.0353
0.7589
0.1191
0.0003
0.0000
tree
0.0200
0.0137
0.1170
0.8493
0.0001
0.0000
car
0.6829
0.0281
0.0053
0.0041
0.2794
0.0002
clutter
0.8667
0.0879
0.0036
0.0418
0.0000
0.0000
Precision/Correctness
0.798
0.920
0.806
0.853
0.845
0.000
Recall/Completeness
0.917
0.948
0.759
0.849
0.279
0.000
F1
0.853
0.934
0.782
0.851
0.420
-nan

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9503
0.0141
0.0262
0.0071
0.0021
0.0001
building
0.0158
0.9683
0.0124
0.0034
0.0000
0.0000
low_veg
0.0762
0.0242
0.7896
0.1096
0.0003
0.0000
tree
0.0101
0.0103
0.1022
0.8773
0.0000
0.0000
car
0.6110
0.0341
0.0027
0.0020
0.3501
0.0001
clutter
0.8912
0.0884
0.0017
0.0186
0.0000
0.0000
Precision/Correctness
0.826
0.942
0.844
0.882
0.871
0.000
Recall/Completeness
0.950
0.968
0.790
0.877
0.350
0.000
F1
0.884
0.955
0.816
0.880
0.500
-nan

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8494
0.1294
0.0102
0.0091
0.0019
0.0000
building
0.0403
0.9462
0.0084
0.0050
0.0002
0.0000
low_veg
0.1533
0.0945
0.4620
0.2899
0.0003
0.0000
tree
0.0459
0.0137
0.0759
0.8643
0.0002
0.0000
car
0.6948
0.0655
0.0041
0.0012
0.2343
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.856
0.924
0.644
0.693
0.834
--
Recall/Completeness
0.849
0.946
0.462
0.864
0.234
--
F1
0.853
0.935
0.538
0.769
0.366
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8802
0.1061
0.0075
0.0051
0.0011
0.0000
building
0.0269
0.9623
0.0072
0.0035
0.0001
0.0000
low_veg
0.1156
0.0652
0.5096
0.3094
0.0002
0.0000
tree
0.0245
0.0042
0.0713
0.9000
0.0000
0.0000
car
0.6531
0.0630
0.0022
0.0000
0.2817
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.897
0.944
0.677
0.727
0.886
--
Recall/Completeness
0.880
0.962
0.510
0.900
0.282
--
F1
0.889
0.953
0.582
0.804
0.427
--

Overall accuracy 0.909

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
0.8667
0.0801
0.0354
0.0143
0.0034
0.0000
building
0.0264
0.9546
0.0099
0.0090
0.0001
0.0000
low_veg
0.0453
0.0797
0.6890
0.1857
0.0004
0.0000
tree
0.0091
0.0062
0.0767
0.9079
0.0002
0.0000
car
0.5591
0.0444
0.0034
0.0020
0.3911
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.897
0.909
0.681
0.908
0.864
--
Recall/Completeness
0.867
0.955
0.689
0.908
0.391
--
F1
0.882
0.931
0.685
0.908
0.538
--

Overall accuracy 0.881

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9079
0.0609
0.0249
0.0047
0.0017
0.0000
building
0.0174
0.9707
0.0065
0.0054
0.0001
0.0000
low_veg
0.0325
0.0636
0.7364
0.1674
0.0001
0.0000
tree
0.0028
0.0029
0.0646
0.9297
0.0001
0.0000
car
0.4889
0.0331
0.0018
0.0000
0.4762
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.934
0.936
0.722
0.935
0.923
--
Recall/Completeness
0.908
0.971
0.736
0.930
0.476
--
F1
0.921
0.953
0.729
0.932
0.628
--

Overall accuracy 0.914

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
0.8782
0.0731
0.0289
0.0172
0.0026
0.0000
building
0.0234
0.9562
0.0117
0.0081
0.0006
0.0000
low_veg
0.0732
0.0241
0.8532
0.0493
0.0002
0.0000
tree
0.0144
0.0110
0.1122
0.8621
0.0003
0.0000
car
0.5646
0.0306
0.0276
0.0056
0.3706
0.0011
clutter
--
--
--
--
--
--
Precision/Correctness
0.894
0.877
0.811
0.924
0.854
--
Recall/Completeness
0.878
0.956
0.853
0.862
0.371
--
F1
0.886
0.915
0.832
0.892
0.517
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9080
0.0655
0.0183
0.0065
0.0017
0.0000
building
0.0131
0.9763
0.0050
0.0051
0.0005
0.0000
low_veg
0.0583
0.0147
0.8877
0.0391
0.0001
0.0000
tree
0.0044
0.0052
0.0948
0.8954
0.0002
0.0000
car
0.4808
0.0325
0.0182
0.0039
0.4628
0.0017
clutter
--
--
--
--
--
--
Precision/Correctness
0.929
0.903
0.852
0.952
0.891
--
Recall/Completeness
0.908
0.976
0.888
0.895
0.463
--
F1
0.918
0.938
0.870
0.923
0.609
--

Overall accuracy 0.912

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
0.8230
0.0863
0.0578
0.0304
0.0026
0.0000
building
0.0139
0.9556
0.0115
0.0183
0.0008
0.0000
low_veg
0.0725
0.0470
0.7372
0.1411
0.0021
0.0001
tree
0.0079
0.0089
0.1097
0.8734
0.0001
0.0000
car
0.4687
0.0639
0.0191
0.0044
0.4435
0.0004
clutter
--
--
--
--
--
--
Precision/Correctness
0.936
0.840
0.734
0.816
0.667
--
Recall/Completeness
0.823
0.956
0.737
0.873
0.443
--
F1
0.876
0.894
0.735
0.844
0.533
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8641
0.0696
0.0483
0.0163
0.0018
0.0000
building
0.0092
0.9704
0.0060
0.0138
0.0007
0.0000
low_veg
0.0710
0.0330
0.7829
0.1107
0.0024
0.0001
tree
0.0036
0.0042
0.0896
0.9026
0.0000
0.0000
car
0.4067
0.0631
0.0127
0.0004
0.5166
0.0006
clutter
--
--
--
--
--
--
Precision/Correctness
0.951
0.875
0.777
0.864
0.685
--
Recall/Completeness
0.864
0.970
0.783
0.903
0.517
--
F1
0.906
0.920
0.780
0.883
0.589
--

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
0.8621
0.0335
0.0677
0.0333
0.0033
0.0001
building
0.1717
0.8096
0.0116
0.0066
0.0005
0.0000
low_veg
0.0611
0.0496
0.7443
0.1418
0.0010
0.0023
tree
0.0079
0.0122
0.2026
0.7769
0.0001
0.0004
car
0.4420
0.0375
0.0166
0.0015
0.5019
0.0004
clutter
--
--
--
--
--
--
Precision/Correctness
0.688
0.866
0.713
0.859
0.745
--
Recall/Completeness
0.862
0.810
0.744
0.777
0.502
--
F1
0.765
0.837
0.728
0.816
0.600
--

Overall accuracy 0.784

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
0.8943
0.0283
0.0573
0.0180
0.0019
0.0001
building
0.1799
0.8125
0.0044
0.0026
0.0005
0.0000
low_veg
0.0601
0.0377
0.7746
0.1241
0.0010
0.0025
tree
0.0037
0.0061
0.1870
0.8029
0.0000
0.0003
car
0.3202
0.0551
0.0139
0.0000
0.6104
0.0005
clutter
--
--
--
--
--
--
Precision/Correctness
0.701
0.902
0.736
0.886
0.776
--
Recall/Completeness
0.894
0.812
0.775
0.803
0.610
--
F1
0.786
0.855
0.755
0.843
0.683
--

Overall accuracy 0.809

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
0.8395
0.0482
0.0699
0.0404
0.0020
0.0000
building
0.0097
0.9643
0.0123
0.0127
0.0009
0.0000
low_veg
0.0708
0.0734
0.7362
0.1179
0.0017
0.0000
tree
0.0194
0.0197
0.1739
0.7870
0.0000
0.0000
car
0.3062
0.0058
0.0055
0.0078
0.6746
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.877
0.823
0.710
0.864
0.629
--
Recall/Completeness
0.839
0.964
0.736
0.787
0.675
--
F1
0.858
0.888
0.723
0.824
0.651
--

Overall accuracy 0.817

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
0.8645
0.0434
0.0623
0.0285
0.0013
0.0000
building
0.0045
0.9827
0.0049
0.0072
0.0007
0.0000
low_veg
0.0664
0.0565
0.7691
0.1064
0.0015
0.0000
tree
0.0132
0.0121
0.1641
0.8107
0.0000
0.0000
car
0.1736
0.0038
0.0021
0.0029
0.8176
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.901
0.858
0.730
0.889
0.668
--
Recall/Completeness
0.865
0.983
0.769
0.811
0.818
--
F1
0.882
0.916
0.749
0.848
0.736
--

Overall accuracy 0.843

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
0.8121
0.0375
0.0918
0.0540
0.0038
0.0008
building
0.0258
0.9420
0.0191
0.0124
0.0003
0.0003
low_veg
0.0516
0.0228
0.8385
0.0846
0.0005
0.0020
tree
0.0221
0.0088
0.2116
0.7566
0.0000
0.0010
car
0.3236
0.0090
0.0455
0.0024
0.6196
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.814
0.914
0.723
0.886
0.632
--
Recall/Completeness
0.812
0.942
0.839
0.757
0.620
--
F1
0.813
0.928
0.776
0.816
0.626
--

Overall accuracy 0.822

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
0.8497
0.0317
0.0798
0.0356
0.0023
0.0007
building
0.0220
0.9616
0.0094
0.0065
0.0003
0.0003
low_veg
0.0487
0.0146
0.8611
0.0731
0.0005
0.0020
tree
0.0157
0.0042
0.1986
0.7806
0.0000
0.0009
car
0.2388
0.0041
0.0338
0.0000
0.7233
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.839
0.942
0.747
0.910
0.667
--
Recall/Completeness
0.850
0.962
0.861
0.781
0.723
--
F1
0.844
0.952
0.800
0.841
0.694
--

Overall accuracy 0.848

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
0.7688
0.0415
0.1390
0.0477
0.0030
0.0000
building
0.0125
0.9466
0.0207
0.0200
0.0002
0.0000
low_veg
0.0588
0.0267
0.7855
0.1269
0.0020
0.0000
tree
0.0046
0.0021
0.1116
0.8817
0.0000
0.0000
car
0.3006
0.0493
0.1180
0.0153
0.5166
0.0000
clutter
0.0247
0.4262
0.2674
0.2817
0.0000
0.0000
Precision/Correctness
0.900
0.856
0.642
0.889
0.673
0.000
Recall/Completeness
0.769
0.947
0.786
0.882
0.517
0.000
F1
0.829
0.899
0.706
0.885
0.585
-nan

Overall accuracy 0.826

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
0.8046
0.0335
0.1304
0.0293
0.0022
0.0000
building
0.0108
0.9609
0.0132
0.0149
0.0002
0.0000
low_veg
0.0546
0.0170
0.8238
0.1023
0.0023
0.0000
tree
0.0017
0.0005
0.0974
0.9005
0.0000
0.0000
car
0.2206
0.0643
0.1039
0.0047
0.6065
0.0000
clutter
0.0210
0.4392
0.2498
0.2900
0.0000
0.0000
Precision/Correctness
0.921
0.876
0.666
0.914
0.669
0.000
Recall/Completeness
0.805
0.961
0.824
0.900
0.606
0.000
F1
0.859
0.916
0.736
0.907
0.636
-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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8588
0.0387
0.0574
0.0413
0.0038
0.0000
building
0.0107
0.9651
0.0133
0.0107
0.0001
0.0000
low_veg
0.0815
0.0600
0.7164
0.1411
0.0010
0.0000
tree
0.0137
0.0060
0.0983
0.8819
0.0002
0.0000
car
0.5609
0.0421
0.0433
0.0076
0.3461
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.906
0.866
0.745
0.853
0.717
--
Recall/Completeness
0.859
0.965
0.716
0.882
0.346
--
F1
0.882
0.913
0.730
0.867
0.467
--

Overall accuracy 0.850

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8971
0.0290
0.0478
0.0231
0.0031
0.0000
building
0.0065
0.9814
0.0060
0.0060
0.0001
0.0000
low_veg
0.0810
0.0447
0.7550
0.1182
0.0011
0.0000
tree
0.0052
0.0023
0.0806
0.9120
0.0000
0.0000
car
0.4799
0.0579
0.0314
0.0017
0.4291
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.929
0.903
0.787
0.892
0.735
--
Recall/Completeness
0.897
0.981
0.755
0.912
0.429
--
F1
0.913
0.940
0.771
0.902
0.542
--

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_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7767
0.0486
0.1368
0.0332
0.0047
0.0000
building
0.0523
0.8929
0.0260
0.0286
0.0002
0.0000
low_veg
0.0262
0.0267
0.8959
0.0501
0.0007
0.0003
tree
0.0112
0.0118
0.2325
0.7442
0.0003
0.0000
car
0.4472
0.0012
0.0098
0.0061
0.5357
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.818
0.889
0.815
0.860
0.650
--
Recall/Completeness
0.777
0.893
0.896
0.744
0.536
--
F1
0.797
0.891
0.853
0.798
0.587
--

Overall accuracy 0.838

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8128
0.0384
0.1228
0.0236
0.0024
0.0000
building
0.0500
0.9052
0.0178
0.0268
0.0002
0.0000
low_veg
0.0233
0.0211
0.9135
0.0409
0.0008
0.0003
tree
0.0069
0.0072
0.2127
0.7730
0.0002
0.0000
car
0.3918
0.0007
0.0032
0.0012
0.6031
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.840
0.915
0.838
0.883
0.686
--
Recall/Completeness
0.813
0.905
0.914
0.773
0.603
--
F1
0.826
0.910
0.874
0.824
0.642
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9248
0.0366
0.0211
0.0104
0.0070
0.0001
building
0.0458
0.9353
0.0091
0.0093
0.0003
0.0002
low_veg
0.1047
0.0424
0.6724
0.1798
0.0001
0.0005
tree
0.0096
0.0083
0.0528
0.9291
0.0002
0.0000
car
0.5551
0.0194
0.0158
0.0087
0.4009
0.0002
clutter
0.7200
0.0975
0.0858
0.0169
0.0798
0.0000
Precision/Correctness
0.838
0.938
0.892
0.689
0.833
0.000
Recall/Completeness
0.925
0.935
0.672
0.929
0.401
0.000
F1
0.879
0.937
0.767
0.791
0.541
-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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9538
0.0247
0.0137
0.0050
0.0028
0.0000
building
0.0350
0.9545
0.0041
0.0060
0.0002
0.0002
low_veg
0.0914
0.0275
0.7160
0.1646
0.0000
0.0006
tree
0.0042
0.0044
0.0385
0.9528
0.0001
0.0000
car
0.4884
0.0183
0.0122
0.0064
0.4747
0.0001
clutter
0.7290
0.0983
0.0700
0.0070
0.0956
0.0000
Precision/Correctness
0.874
0.961
0.931
0.726
0.907
0.000
Recall/Completeness
0.954
0.955
0.716
0.953
0.475
0.000
F1
0.912
0.958
0.809
0.824
0.623
-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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9373
0.0214
0.0218
0.0158
0.0037
0.0001
building
0.0781
0.9042
0.0078
0.0099
0.0001
0.0000
low_veg
0.0626
0.0407
0.6372
0.2584
0.0009
0.0002
tree
0.0046
0.0085
0.0296
0.9573
0.0000
0.0000
car
0.4973
0.0190
0.0038
0.0151
0.4647
0.0000
clutter
0.5844
0.2912
0.0239
0.0091
0.0877
0.0038
Precision/Correctness
0.878
0.908
0.860
0.686
0.591
0.665
Recall/Completeness
0.937
0.904
0.637
0.957
0.465
0.004
F1
0.907
0.906
0.732
0.799
0.520
0.007

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
0.9600
0.0142
0.0159
0.0078
0.0020
0.0001
building
0.0739
0.9163
0.0034
0.0063
0.0000
0.0000
low_veg
0.0404
0.0247
0.6912
0.2429
0.0006
0.0002
tree
0.0019
0.0065
0.0218
0.9699
0.0000
0.0000
car
0.4287
0.0162
0.0016
0.0115
0.5419
0.0000
clutter
0.5705
0.3117
0.0176
0.0023
0.0936
0.0043
Precision/Correctness
0.900
0.932
0.899
0.728
0.616
0.674
Recall/Completeness
0.960
0.916
0.691
0.970
0.542
0.004
F1
0.929
0.924
0.781
0.832
0.576
0.009

Overall accuracy 0.881

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9688
0.0088
0.0086
0.0067
0.0071
0.0000
building
0.3288
0.6558
0.0072
0.0035
0.0047
0.0000
low_veg
0.0700
0.0033
0.9006
0.0250
0.0009
0.0002
tree
0.0267
0.0032
0.1119
0.8579
0.0002
0.0000
car
0.4641
0.1103
0.0088
0.0086
0.4076
0.0007
clutter
0.5276
0.3347
0.0391
0.0070
0.0843
0.0073
Precision/Correctness
0.731
0.903
0.872
0.926
0.675
0.809
Recall/Completeness
0.969
0.656
0.901
0.858
0.408
0.007
F1
0.833
0.760
0.886
0.891
0.508
0.014

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9823
0.0065
0.0059
0.0016
0.0037
0.0000
building
0.3241
0.6682
0.0025
0.0012
0.0040
0.0000
low_veg
0.0436
0.0011
0.9378
0.0169
0.0004
0.0001
tree
0.0107
0.0005
0.0959
0.8929
0.0000
0.0000
car
0.3835
0.1387
0.0042
0.0023
0.4702
0.0010
clutter
0.4706
0.3906
0.0361
0.0023
0.0924
0.0079
Precision/Correctness
0.751
0.913
0.905
0.964
0.712
0.864
Recall/Completeness
0.982
0.668
0.938
0.893
0.470
0.008
F1
0.851
0.772
0.921
0.927
0.566
0.016

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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9756
0.0110
0.0076
0.0039
0.0019
0.0000
building
0.2161
0.7014
0.0087
0.0712
0.0026
0.0000
low_veg
0.0843
0.0138
0.8659
0.0355
0.0005
0.0000
tree
0.0296
0.0148
0.1653
0.7904
0.0000
0.0000
car
0.5580
0.0001
0.0154
0.0024
0.4242
0.0000
clutter
0.6638
0.1814
0.0077
0.0015
0.0764
0.0693
Precision/Correctness
0.755
0.920
0.851
0.749
0.489
0.999
Recall/Completeness
0.976
0.701
0.866
0.790
0.424
0.069
F1
0.851
0.796
0.858
0.769
0.454
0.130

Overall accuracy 0.804

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
0.9880
0.0069
0.0031
0.0009
0.0010
0.0000
building
0.2140
0.7088
0.0036
0.0712
0.0024
0.0000
low_veg
0.0510
0.0100
0.9062
0.0325
0.0004
0.0000
tree
0.0109
0.0117
0.1548
0.8226
0.0000
0.0000
car
0.4803
0.0001
0.0107
0.0008
0.5082
0.0000
clutter
0.6327
0.1947
0.0059
0.0006
0.0859
0.0803
Precision/Correctness
0.776
0.934
0.881
0.759
0.479
1.000
Recall/Completeness
0.988
0.709
0.906
0.823
0.508
0.080
F1
0.869
0.806
0.893
0.790
0.493
0.149

Overall accuracy 0.824

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
0.8520
0.0558
0.0644
0.0258
0.0013
0.0007
building
0.0193
0.9590
0.0157
0.0052
0.0007
0.0001
low_veg
0.0597
0.0289
0.7288
0.1819
0.0006
0.0000
tree
0.0188
0.0087
0.1734
0.7990
0.0001
0.0000
car
0.6284
0.0871
0.0097
0.0128
0.2438
0.0182
clutter
--
--
--
--
--
--
Precision/Correctness
0.880
0.914
0.748
0.779
0.734
--
Recall/Completeness
0.852
0.959
0.729
0.799
0.244
--
F1
0.866
0.936
0.738
0.789
0.366
--

Overall accuracy 0.833

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
0.9089
0.0455
0.0351
0.0089
0.0010
0.0007
building
0.0119
0.9807
0.0054
0.0014
0.0005
0.0001
low_veg
0.0495
0.0168
0.7598
0.1733
0.0006
0.0000
tree
0.0069
0.0034
0.1424
0.8472
0.0000
0.0000
car
0.5553
0.1065
0.0045
0.0054
0.3082
0.0201
clutter
--
--
--
--
--
--
Precision/Correctness
0.917
0.943
0.813
0.811
0.767
--
Recall/Completeness
0.909
0.981
0.760
0.847
0.308
--
F1
0.913
0.961
0.786
0.829
0.440
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8927
0.0389
0.0434
0.0186
0.0063
0.0001
building
0.0258
0.9366
0.0222
0.0151
0.0003
0.0000
low_veg
0.0516
0.0276
0.8537
0.0665
0.0006
0.0000
tree
0.0155
0.0097
0.1587
0.8160
0.0001
0.0000
car
0.5017
0.0257
0.0600
0.0087
0.4032
0.0007
clutter
--
--
--
--
--
--
Precision/Correctness
0.893
0.921
0.830
0.853
0.665
--
Recall/Completeness
0.893
0.937
0.854
0.816
0.403
--
F1
0.893
0.929
0.842
0.834
0.502
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9261
0.0347
0.0272
0.0065
0.0053
0.0002
building
0.0165
0.9629
0.0100
0.0104
0.0002
0.0000
low_veg
0.0382
0.0199
0.8903
0.0511
0.0005
0.0000
tree
0.0069
0.0065
0.1334
0.8532
0.0000
0.0000
car
0.4220
0.0299
0.0495
0.0034
0.4946
0.0005
clutter
--
--
--
--
--
--
Precision/Correctness
0.928
0.940
0.874
0.897
0.690
--
Recall/Completeness
0.926
0.963
0.890
0.853
0.495
--
F1
0.927
0.951
0.882
0.874
0.576
--

Overall accuracy 0.909

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image