Detailled semantic labeling (2D) result


Name K. Nogueira et al.
Affiliation Universidade Federal de Minas Gerais, Brazil
Abbreviation UFMG_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
0.902
0.041
0.039
0.013
0.004
0.002
building
0.050
0.925
0.017
0.007
0.001
0.000
low_veg
0.072
0.026
0.761
0.139
0.001
0.001
tree
0.027
0.008
0.091
0.873
0.000
0.000
car
0.302
0.033
0.014
0.006
0.645
0.000
clutter
0.512
0.175
0.021
0.005
0.056
0.230
Precision/Correctness
0.854
0.920
0.820
0.847
0.806
0.693
Recall/Completeness
0.902
0.925
0.761
0.873
0.645
0.230
F1
0.877
0.922
0.790
0.860
0.717
0.345

Overall accuracy 0.862

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
0.927
0.032
0.031
0.006
0.002
0.002
building
0.044
0.939
0.011
0.005
0.001
0.000
low_veg
0.059
0.016
0.798
0.126
0.000
0.001
tree
0.014
0.004
0.075
0.907
0.000
0.000
car
0.206
0.035
0.011
0.003
0.745
0.000
clutter
0.492
0.178
0.018
0.002
0.057
0.253
Precision/Correctness
0.885
0.944
0.854
0.873
0.829
0.721
Recall/Completeness
0.927
0.939
0.798
0.907
0.745
0.253
F1
0.905
0.941
0.825
0.890
0.785
0.375

Overall accuracy 0.891
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9479
0.0139
0.0252
0.0084
0.0029
0.0015
building
0.0373
0.9430
0.0166
0.0026
0.0004
0.0002
low_veg
0.1071
0.0278
0.7288
0.1356
0.0006
0.0002
tree
0.0392
0.0117
0.0919
0.8556
0.0001
0.0016
car
0.3932
0.0017
0.0009
0.0017
0.6026
0.0000
clutter
0.0061
0.0000
0.0000
0.0097
0.0000
0.9842
Precision/Correctness
0.824
0.943
0.845
0.854
0.904
0.955
Recall/Completeness
0.948
0.943
0.729
0.856
0.603
0.984
F1
0.882
0.943
0.782
0.855
0.723
0.969

Overall accuracy 0.866

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
0.9702
0.0089
0.0150
0.0022
0.0022
0.0015
building
0.0301
0.9592
0.0092
0.0009
0.0004
0.0002
low_veg
0.0901
0.0178
0.7607
0.1306
0.0006
0.0002
tree
0.0209
0.0084
0.0784
0.8916
0.0001
0.0006
car
0.2640
0.0007
0.0004
0.0004
0.7345
0.0000
clutter
0.0061
0.0000
0.0000
0.0021
0.0000
0.9917
Precision/Correctness
0.865
0.964
0.877
0.874
0.918
0.969
Recall/Completeness
0.970
0.959
0.761
0.892
0.734
0.992
F1
0.915
0.961
0.815
0.883
0.816
0.980

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8824
0.0990
0.0092
0.0040
0.0036
0.0018
building
0.0418
0.9500
0.0069
0.0012
0.0000
0.0000
low_veg
0.1753
0.0724
0.4888
0.2592
0.0025
0.0017
tree
0.0620
0.0266
0.0718
0.8389
0.0001
0.0006
car
0.4129
0.0218
0.0026
0.0009
0.5618
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.865
0.941
0.685
0.750
0.873
--
Recall/Completeness
0.882
0.950
0.489
0.839
0.562
--
F1
0.874
0.945
0.570
0.792
0.684
--

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
0.9125
0.0741
0.0067
0.0022
0.0024
0.0020
building
0.0312
0.9622
0.0059
0.0007
0.0000
0.0000
low_veg
0.1376
0.0375
0.5321
0.2882
0.0025
0.0021
tree
0.0309
0.0116
0.0650
0.8919
0.0000
0.0005
car
0.3147
0.0174
0.0023
0.0000
0.6656
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.900
0.960
0.716
0.769
0.900
--
Recall/Completeness
0.913
0.962
0.532
0.892
0.666
--
F1
0.906
0.961
0.611
0.826
0.765
--

Overall accuracy 0.923

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
0.8926
0.0689
0.0272
0.0055
0.0051
0.0007
building
0.0340
0.9472
0.0160
0.0025
0.0001
0.0003
low_veg
0.1015
0.0454
0.7031
0.1496
0.0003
0.0001
tree
0.0214
0.0063
0.0532
0.9188
0.0001
0.0002
car
0.3346
0.0273
0.0010
0.0007
0.6364
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.871
0.927
0.731
0.936
0.880
--
Recall/Completeness
0.893
0.947
0.703
0.919
0.636
--
F1
0.882
0.937
0.717
0.927
0.739
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9270
0.0496
0.0185
0.0013
0.0032
0.0006
building
0.0271
0.9597
0.0121
0.0008
0.0000
0.0003
low_veg
0.0809
0.0307
0.7529
0.1354
0.0000
0.0001
tree
0.0084
0.0026
0.0410
0.9479
0.0001
0.0001
car
0.2327
0.0188
0.0004
0.0000
0.7480
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.911
0.953
0.778
0.953
0.918
--
Recall/Completeness
0.927
0.960
0.753
0.948
0.748
--
F1
0.919
0.956
0.765
0.951
0.824
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8923
0.0426
0.0503
0.0065
0.0026
0.0056
building
0.0260
0.9583
0.0126
0.0021
0.0003
0.0008
low_veg
0.0719
0.0197
0.8521
0.0550
0.0003
0.0010
tree
0.0317
0.0089
0.0893
0.8685
0.0001
0.0015
car
0.3476
0.0216
0.0187
0.0033
0.6082
0.0005
clutter
--
--
--
--
--
--
Precision/Correctness
0.893
0.918
0.806
0.939
0.912
--
Recall/Completeness
0.892
0.958
0.852
0.869
0.608
--
F1
0.893
0.938
0.828
0.903
0.730
--

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
0.9158
0.0349
0.0410
0.0020
0.0014
0.0049
building
0.0170
0.9757
0.0056
0.0006
0.0003
0.0008
low_veg
0.0548
0.0107
0.8883
0.0456
0.0001
0.0007
tree
0.0135
0.0043
0.0703
0.9109
0.0000
0.0009
car
0.2462
0.0237
0.0123
0.0025
0.7151
0.0001
clutter
--
--
--
--
--
--
Precision/Correctness
0.931
0.942
0.847
0.958
0.944
--
Recall/Completeness
0.916
0.976
0.888
0.911
0.715
--
F1
0.924
0.959
0.867
0.934
0.814
--

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
0.8489
0.0874
0.0438
0.0152
0.0042
0.0005
building
0.0136
0.9596
0.0148
0.0099
0.0020
0.0000
low_veg
0.0796
0.0373
0.7250
0.1567
0.0014
0.0001
tree
0.0184
0.0106
0.0695
0.9015
0.0000
0.0000
car
0.3121
0.0374
0.0069
0.0034
0.6402
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.932
0.844
0.788
0.839
0.665
--
Recall/Completeness
0.849
0.960
0.725
0.901
0.640
--
F1
0.889
0.898
0.755
0.869
0.652
--

Overall accuracy 0.863

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
0.8815
0.0728
0.0365
0.0055
0.0031
0.0006
building
0.0100
0.9702
0.0108
0.0072
0.0018
0.0000
low_veg
0.0747
0.0221
0.7744
0.1272
0.0015
0.0001
tree
0.0083
0.0050
0.0530
0.9337
0.0000
0.0000
car
0.2388
0.0379
0.0058
0.0007
0.7168
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.950
0.877
0.828
0.880
0.667
--
Recall/Completeness
0.881
0.970
0.774
0.934
0.717
--
F1
0.914
0.921
0.800
0.906
0.691
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8680
0.0385
0.0596
0.0302
0.0021
0.0015
building
0.1754
0.8072
0.0130
0.0033
0.0012
0.0000
low_veg
0.0603
0.0425
0.6917
0.2037
0.0008
0.0010
tree
0.0121
0.0101
0.0918
0.8860
0.0000
0.0000
car
0.2463
0.0029
0.0067
0.0130
0.7312
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.685
0.878
0.817
0.837
0.829
--
Recall/Completeness
0.868
0.807
0.692
0.886
0.731
--
F1
0.766
0.841
0.749
0.861
0.777
--

Overall accuracy 0.811

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
0.8965
0.0333
0.0515
0.0167
0.0004
0.0016
building
0.1844
0.8060
0.0074
0.0009
0.0012
0.0000
low_veg
0.0584
0.0318
0.7273
0.1806
0.0008
0.0011
tree
0.0061
0.0042
0.0789
0.9109
0.0000
0.0000
car
0.1465
0.0017
0.0068
0.0072
0.8378
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.699
0.911
0.844
0.863
0.857
--
Recall/Completeness
0.897
0.806
0.727
0.911
0.838
--
F1
0.785
0.855
0.781
0.886
0.847
--

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8469
0.0596
0.0608
0.0282
0.0037
0.0008
building
0.0253
0.9537
0.0168
0.0038
0.0004
0.0000
low_veg
0.1048
0.0551
0.7039
0.1353
0.0008
0.0001
tree
0.0380
0.0137
0.1253
0.8228
0.0001
0.0000
car
0.1635
0.0000
0.0005
0.0033
0.8327
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.814
0.838
0.752
0.869
0.664
--
Recall/Completeness
0.847
0.954
0.704
0.823
0.833
--
F1
0.830
0.892
0.727
0.845
0.739
--

Overall accuracy 0.821

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
0.8660
0.0560
0.0555
0.0189
0.0029
0.0007
building
0.0198
0.9702
0.0083
0.0014
0.0003
0.0000
low_veg
0.0992
0.0396
0.7363
0.1241
0.0007
0.0000
tree
0.0283
0.0067
0.1146
0.8503
0.0001
0.0000
car
0.0605
0.0000
0.0000
0.0003
0.9391
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.842
0.870
0.775
0.890
0.668
--
Recall/Completeness
0.866
0.970
0.736
0.850
0.939
--
F1
0.854
0.918
0.755
0.870
0.781
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7815
0.0407
0.1164
0.0471
0.0020
0.0123
building
0.0276
0.9405
0.0201
0.0089
0.0000
0.0028
low_veg
0.0205
0.0167
0.8048
0.1571
0.0001
0.0008
tree
0.0207
0.0067
0.1187
0.8522
0.0000
0.0017
car
0.1537
0.0127
0.0338
0.0132
0.7866
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.864
0.924
0.785
0.849
0.830
--
Recall/Completeness
0.782
0.940
0.805
0.852
0.787
--
F1
0.821
0.932
0.795
0.851
0.808
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8169
0.0356
0.1062
0.0281
0.0006
0.0125
building
0.0241
0.9595
0.0100
0.0037
0.0000
0.0027
low_veg
0.0171
0.0099
0.8290
0.1431
0.0002
0.0007
tree
0.0137
0.0025
0.1064
0.8760
0.0000
0.0014
car
0.0927
0.0074
0.0214
0.0055
0.8730
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.894
0.949
0.812
0.872
0.906
--
Recall/Completeness
0.817
0.959
0.829
0.876
0.873
--
F1
0.854
0.954
0.820
0.874
0.889
--

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8449
0.0276
0.0905
0.0328
0.0041
0.0000
building
0.0159
0.9458
0.0321
0.0062
0.0000
0.0000
low_veg
0.0739
0.0210
0.7461
0.1585
0.0006
0.0000
tree
0.0103
0.0023
0.0493
0.9381
0.0000
0.0000
car
0.1988
0.0119
0.0359
0.0113
0.7420
0.0000
clutter
0.9586
0.0200
0.0207
0.0005
0.0002
0.0000
Precision/Correctness
0.780
0.933
0.765
0.909
0.764
0.000
Recall/Completeness
0.845
0.946
0.746
0.938
0.742
0.000
F1
0.811
0.939
0.755
0.924
0.753
-nan

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8732
0.0192
0.0865
0.0181
0.0030
0.0000
building
0.0137
0.9557
0.0273
0.0033
0.0000
0.0000
low_veg
0.0673
0.0097
0.7930
0.1294
0.0005
0.0000
tree
0.0044
0.0005
0.0376
0.9574
0.0000
0.0000
car
0.1117
0.0091
0.0287
0.0038
0.8466
0.0000
clutter
0.9684
0.0177
0.0136
0.0001
0.0002
0.0000
Precision/Correctness
0.798
0.961
0.794
0.934
0.779
-nan
Recall/Completeness
0.873
0.956
0.793
0.957
0.847
0.000
F1
0.834
0.958
0.793
0.946
0.811
-nan

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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9003
0.0286
0.0421
0.0236
0.0051
0.0002
building
0.0376
0.9465
0.0112
0.0042
0.0005
0.0000
low_veg
0.1197
0.0398
0.7048
0.1350
0.0005
0.0001
tree
0.0262
0.0052
0.0625
0.9059
0.0002
0.0001
car
0.3180
0.0140
0.0286
0.0078
0.6316
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.873
0.900
0.808
0.880
0.788
--
Recall/Completeness
0.900
0.947
0.705
0.906
0.632
--
F1
0.886
0.922
0.753
0.893
0.701
--

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
0.9289
0.0212
0.0357
0.0104
0.0036
0.0002
building
0.0327
0.9592
0.0057
0.0021
0.0003
0.0000
low_veg
0.1164
0.0246
0.7495
0.1089
0.0004
0.0001
tree
0.0110
0.0019
0.0459
0.9412
0.0000
0.0001
car
0.2254
0.0182
0.0243
0.0020
0.7301
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.900
0.934
0.849
0.915
0.819
--
Recall/Completeness
0.929
0.959
0.750
0.941
0.730
--
F1
0.914
0.946
0.796
0.928
0.772
--

Overall accuracy 0.901

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
0.8177
0.0430
0.1034
0.0296
0.0045
0.0018
building
0.0606
0.8953
0.0281
0.0158
0.0002
0.0001
low_veg
0.0337
0.0194
0.8763
0.0686
0.0004
0.0017
tree
0.0235
0.0103
0.1497
0.8164
0.0000
0.0000
car
0.2196
0.0001
0.0053
0.0060
0.7690
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.788
0.909
0.861
0.854
0.775
--
Recall/Completeness
0.818
0.895
0.876
0.816
0.769
--
F1
0.802
0.902
0.868
0.835
0.772
--

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
0.8498
0.0332
0.0924
0.0199
0.0030
0.0017
building
0.0578
0.9062
0.0214
0.0143
0.0002
0.0001
low_veg
0.0305
0.0141
0.8962
0.0570
0.0004
0.0018
tree
0.0168
0.0062
0.1297
0.8472
0.0000
0.0000
car
0.1162
0.0000
0.0012
0.0012
0.8814
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.811
0.934
0.883
0.878
0.790
--
Recall/Completeness
0.850
0.906
0.896
0.847
0.881
--
F1
0.830
0.920
0.889
0.862
0.833
--

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
0.9336
0.0337
0.0189
0.0054
0.0055
0.0030
building
0.0529
0.9275
0.0130
0.0050
0.0004
0.0012
low_veg
0.1085
0.0275
0.6772
0.1856
0.0004
0.0009
tree
0.0195
0.0078
0.0398
0.9321
0.0001
0.0007
car
0.3219
0.0043
0.0084
0.0061
0.6593
0.0000
clutter
0.7118
0.1090
0.0764
0.0135
0.0886
0.0007
Precision/Correctness
0.847
0.949
0.898
0.700
0.907
0.001
Recall/Completeness
0.934
0.927
0.677
0.932
0.659
0.001
F1
0.888
0.938
0.772
0.800
0.764
0.001

Overall accuracy 0.870

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
0.9574
0.0231
0.0132
0.0019
0.0014
0.0030
building
0.0418
0.9459
0.0078
0.0031
0.0003
0.0012
low_veg
0.0896
0.0148
0.7185
0.1761
0.0003
0.0007
tree
0.0098
0.0031
0.0250
0.9617
0.0000
0.0003
car
0.2240
0.0026
0.0067
0.0051
0.7615
0.0000
clutter
0.7196
0.1058
0.0649
0.0071
0.1022
0.0004
Precision/Correctness
0.884
0.970
0.933
0.728
0.959
0.000
Recall/Completeness
0.957
0.946
0.718
0.962
0.761
0.000
F1
0.919
0.958
0.812
0.828
0.849
0.000

Overall accuracy 0.901

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
0.9380
0.0218
0.0203
0.0128
0.0052
0.0019
building
0.0508
0.9350
0.0072
0.0055
0.0015
0.0000
low_veg
0.0757
0.0298
0.5927
0.3006
0.0007
0.0006
tree
0.0070
0.0086
0.0182
0.9662
0.0000
0.0000
car
0.2991
0.0051
0.0085
0.0074
0.6798
0.0000
clutter
0.6084
0.2667
0.0166
0.0089
0.0797
0.0197
Precision/Correctness
0.893
0.918
0.870
0.672
0.642
0.392
Recall/Completeness
0.938
0.935
0.593
0.966
0.680
0.020
F1
0.915
0.927
0.705
0.792
0.660
0.038

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9599
0.0137
0.0146
0.0060
0.0039
0.0020
building
0.0472
0.9450
0.0036
0.0027
0.0014
0.0000
low_veg
0.0511
0.0139
0.6402
0.2936
0.0006
0.0006
tree
0.0019
0.0057
0.0125
0.9798
0.0000
0.0000
car
0.2118
0.0011
0.0081
0.0042
0.7748
0.0000
clutter
0.6126
0.2675
0.0101
0.0032
0.0849
0.0217
Precision/Correctness
0.914
0.944
0.907
0.705
0.641
0.385
Recall/Completeness
0.960
0.945
0.640
0.980
0.775
0.022
F1
0.936
0.945
0.751
0.820
0.702
0.041

Overall accuracy 0.886

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9597
0.0161
0.0111
0.0040
0.0070
0.0021
building
0.0529
0.9265
0.0114
0.0019
0.0068
0.0003
low_veg
0.0507
0.0038
0.8907
0.0531
0.0005
0.0013
tree
0.0430
0.0054
0.0589
0.8922
0.0002
0.0003
car
0.1761
0.1167
0.0266
0.0092
0.6713
0.0001
clutter
0.4220
0.3675
0.0525
0.0069
0.0574
0.0937
Precision/Correctness
0.887
0.914
0.887
0.906
0.785
0.734
Recall/Completeness
0.960
0.927
0.891
0.892
0.671
0.094
F1
0.922
0.920
0.889
0.899
0.724
0.166

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9742
0.0112
0.0083
0.0009
0.0033
0.0020
building
0.0457
0.9397
0.0078
0.0003
0.0062
0.0003
low_veg
0.0302
0.0009
0.9238
0.0440
0.0001
0.0010
tree
0.0151
0.0012
0.0454
0.9382
0.0000
0.0001
car
0.1095
0.1246
0.0217
0.0037
0.7405
0.0000
clutter
0.3646
0.4128
0.0555
0.0023
0.0588
0.1060
Precision/Correctness
0.919
0.929
0.915
0.934
0.814
0.756
Recall/Completeness
0.974
0.940
0.924
0.938
0.740
0.106
F1
0.946
0.935
0.920
0.936
0.776
0.186

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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9709
0.0148
0.0100
0.0016
0.0022
0.0005
building
0.1469
0.7723
0.0212
0.0535
0.0037
0.0025
low_veg
0.0683
0.0166
0.7998
0.1138
0.0015
0.0000
tree
0.0541
0.0142
0.1384
0.7933
0.0000
0.0000
car
0.2944
0.0201
0.0081
0.0034
0.6740
0.0000
clutter
0.5516
0.1523
0.0079
0.0001
0.1003
0.1878
Precision/Correctness
0.802
0.925
0.831
0.712
0.534
0.917
Recall/Completeness
0.971
0.772
0.800
0.793
0.674
0.188
F1
0.878
0.842
0.815
0.750
0.596
0.312

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_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9826
0.0101
0.0058
0.0002
0.0011
0.0003
building
0.1443
0.7773
0.0184
0.0542
0.0032
0.0025
low_veg
0.0373
0.0112
0.8309
0.1191
0.0014
0.0000
tree
0.0253
0.0099
0.1267
0.8381
0.0000
0.0000
car
0.1829
0.0229
0.0066
0.0020
0.7855
0.0001
clutter
0.5183
0.1581
0.0077
0.0000
0.1071
0.2088
Precision/Correctness
0.826
0.941
0.856
0.714
0.529
0.921
Recall/Completeness
0.983
0.777
0.831
0.838
0.785
0.209
F1
0.898
0.851
0.843
0.771
0.632
0.340

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8953
0.0376
0.0550
0.0092
0.0008
0.0020
building
0.0611
0.9175
0.0176
0.0023
0.0008
0.0007
low_veg
0.0843
0.0170
0.7151
0.1827
0.0005
0.0004
tree
0.0472
0.0065
0.1253
0.8201
0.0000
0.0008
car
0.3620
0.0322
0.0017
0.0067
0.5973
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.815
0.941
0.788
0.799
0.890
--
Recall/Completeness
0.895
0.917
0.715
0.820
0.597
--
F1
0.853
0.929
0.750
0.809
0.715
--

Overall accuracy 0.837

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
0.9295
0.0306
0.0359
0.0020
0.0004
0.0017
building
0.0498
0.9408
0.0078
0.0003
0.0007
0.0006
low_veg
0.0640
0.0094
0.7455
0.1802
0.0005
0.0004
tree
0.0173
0.0021
0.0987
0.8814
0.0000
0.0004
car
0.2495
0.0419
0.0002
0.0019
0.7065
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.867
0.962
0.845
0.818
0.912
--
Recall/Completeness
0.930
0.941
0.746
0.881
0.707
--
F1
0.897
0.951
0.792
0.849
0.796
--

Overall accuracy 0.876

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9024
0.0358
0.0487
0.0095
0.0028
0.0009
building
0.0248
0.9342
0.0353
0.0053
0.0005
0.0000
low_veg
0.0444
0.0147
0.8356
0.1051
0.0003
0.0000
tree
0.0315
0.0070
0.1072
0.8543
0.0001
0.0000
car
0.3024
0.0549
0.0366
0.0092
0.5968
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.897
0.938
0.844
0.833
0.856
--
Recall/Completeness
0.902
0.934
0.836
0.854
0.597
--
F1
0.900
0.936
0.840
0.844
0.703
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9285
0.0322
0.0336
0.0030
0.0017
0.0010
building
0.0182
0.9558
0.0229
0.0026
0.0005
0.0000
low_veg
0.0293
0.0078
0.8719
0.0907
0.0002
0.0000
tree
0.0137
0.0037
0.0853
0.8972
0.0001
0.0000
car
0.2141
0.0616
0.0239
0.0050
0.6954
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.936
0.955
0.885
0.865
0.885
--
Recall/Completeness
0.929
0.956
0.872
0.897
0.695
--
F1
0.932
0.955
0.878
0.881
0.779
--

Overall accuracy 0.913

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image