Detailled semantic labeling (2D) result


Name Paisitkriangkrai, Sherrah, Janney, van den Hengel (CNN only)
Affiliation Uni of Adelaide & DSTO,Australia
Abbreviation ADL_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
0.855
0.040
0.053
0.023
0.029
0.000
building
0.070
0.884
0.029
0.012
0.006
0.000
low_veg
0.063
0.027
0.728
0.178
0.004
0.000
tree
0.025
0.007
0.096
0.871
0.001
0.000
car
0.161
0.019
0.018
0.008
0.793
0.000
clutter
0.448
0.270
0.044
0.017
0.220
0.000
Precision/Correctness
0.846
0.915
0.780
0.803
0.452
-nan
Recall/Completeness
0.855
0.884
0.728
0.871
0.793
0.000
F1
0.850
0.899
0.753
0.836
0.576
-nan

Overall accuracy 0.830

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
0.891
0.032
0.042
0.013
0.022
0.000
building
0.063
0.903
0.021
0.009
0.005
0.000
low_veg
0.052
0.016
0.764
0.165
0.003
0.000
tree
0.015
0.003
0.080
0.901
0.001
0.000
car
0.111
0.018
0.013
0.005
0.853
0.000
clutter
0.429
0.289
0.040
0.010
0.232
0.000
Precision/Correctness
0.872
0.939
0.819
0.832
0.451
-nan
Recall/Completeness
0.891
0.903
0.764
0.901
0.853
0.000
F1
0.881
0.920
0.790
0.865
0.590
-nan

Overall accuracy 0.861
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8746
0.0129
0.0658
0.0173
0.0294
0.0000
building
0.0376
0.9118
0.0392
0.0088
0.0026
0.0000
low_veg
0.0728
0.0190
0.7931
0.1118
0.0033
0.0000
tree
0.0227
0.0072
0.1307
0.8382
0.0011
0.0000
car
0.2135
0.0030
0.0088
0.0044
0.7703
0.0000
clutter
0.9159
0.0244
0.0073
0.0513
0.0011
0.0000
Precision/Correctness
0.807
0.954
0.770
0.856
0.570
-nan
Recall/Completeness
0.875
0.912
0.793
0.838
0.770
0.000
F1
0.839
0.932
0.781
0.847
0.655
-nan

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9137
0.0086
0.0483
0.0075
0.0219
0.0000
building
0.0298
0.9368
0.0257
0.0057
0.0021
0.0000
low_veg
0.0613
0.0090
0.8233
0.1038
0.0025
0.0000
tree
0.0115
0.0035
0.1143
0.8701
0.0006
0.0000
car
0.1422
0.0022
0.0069
0.0013
0.8474
0.0000
clutter
0.9417
0.0248
0.0051
0.0278
0.0006
0.0000
Precision/Correctness
0.834
0.975
0.809
0.883
0.604
-nan
Recall/Completeness
0.914
0.937
0.823
0.870
0.847
0.000
F1
0.872
0.956
0.816
0.877
0.705
-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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8294
0.0963
0.0274
0.0083
0.0386
0.0000
building
0.0538
0.9263
0.0132
0.0026
0.0042
0.0000
low_veg
0.1563
0.0841
0.5214
0.2300
0.0081
0.0000
tree
0.0661
0.0152
0.0951
0.8191
0.0045
0.0000
car
0.2003
0.0390
0.0144
0.0055
0.7408
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.848
0.940
0.536
0.734
0.431
--
Recall/Completeness
0.829
0.926
0.521
0.819
0.741
--
F1
0.839
0.933
0.528
0.774
0.545
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8707
0.0683
0.0231
0.0049
0.0330
0.0000
building
0.0426
0.9413
0.0110
0.0014
0.0036
0.0000
low_veg
0.1297
0.0485
0.5704
0.2464
0.0050
0.0000
tree
0.0439
0.0067
0.0882
0.8595
0.0017
0.0000
car
0.1563
0.0295
0.0136
0.0036
0.7970
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.879
0.962
0.568
0.769
0.427
--
Recall/Completeness
0.871
0.941
0.570
0.859
0.797
--
F1
0.875
0.951
0.569
0.812
0.556
--

Overall accuracy 0.900

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
0.8306
0.0614
0.0545
0.0125
0.0410
0.0000
building
0.0473
0.9106
0.0263
0.0116
0.0043
0.0000
low_veg
0.0729
0.0385
0.7111
0.1737
0.0038
0.0000
tree
0.0160
0.0045
0.0731
0.9060
0.0004
0.0000
car
0.1604
0.0238
0.0113
0.0048
0.7997
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.874
0.934
0.634
0.911
0.514
--
Recall/Completeness
0.831
0.911
0.711
0.906
0.800
--
F1
0.852
0.922
0.670
0.908
0.626
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8783
0.0434
0.0425
0.0047
0.0311
0.0000
building
0.0395
0.9303
0.0190
0.0082
0.0030
0.0000
low_veg
0.0608
0.0238
0.7572
0.1555
0.0028
0.0000
tree
0.0080
0.0024
0.0583
0.9312
0.0001
0.0000
car
0.1144
0.0144
0.0078
0.0009
0.8625
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.904
0.958
0.681
0.935
0.538
--
Recall/Completeness
0.878
0.930
0.757
0.931
0.862
--
F1
0.891
0.944
0.717
0.933
0.663
--

Overall accuracy 0.900

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
0.8350
0.0379
0.0824
0.0172
0.0275
0.0000
building
0.0366
0.9264
0.0277
0.0065
0.0028
0.0000
low_veg
0.0738
0.0172
0.8471
0.0594
0.0025
0.0000
tree
0.0296
0.0071
0.1062
0.8561
0.0010
0.0000
car
0.1776
0.0072
0.0310
0.0089
0.7753
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.888
0.926
0.741
0.916
0.567
--
Recall/Completeness
0.835
0.926
0.847
0.856
0.775
--
F1
0.861
0.926
0.791
0.885
0.655
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8723
0.0315
0.0688
0.0072
0.0202
0.0000
building
0.0262
0.9486
0.0192
0.0038
0.0022
0.0000
low_veg
0.0598
0.0096
0.8816
0.0474
0.0016
0.0000
tree
0.0179
0.0037
0.0865
0.8913
0.0006
0.0000
car
0.1341
0.0058
0.0191
0.0066
0.8343
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.919
0.947
0.784
0.945
0.594
--
Recall/Completeness
0.872
0.949
0.882
0.891
0.834
--
F1
0.895
0.948
0.830
0.917
0.694
--

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_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7753
0.1166
0.0673
0.0255
0.0153
0.0000
building
0.0232
0.9233
0.0309
0.0187
0.0040
0.0000
low_veg
0.0688
0.0355
0.7227
0.1706
0.0024
0.0000
tree
0.0225
0.0066
0.0851
0.8855
0.0003
0.0000
car
0.1406
0.0394
0.0086
0.0109
0.8006
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.926
0.809
0.717
0.805
0.441
--
Recall/Completeness
0.775
0.923
0.723
0.886
0.801
--
F1
0.844
0.862
0.720
0.843
0.569
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8098
0.1072
0.0565
0.0142
0.0123
0.0000
building
0.0161
0.9441
0.0220
0.0144
0.0034
0.0000
low_veg
0.0654
0.0238
0.7669
0.1416
0.0024
0.0000
tree
0.0134
0.0035
0.0642
0.9188
0.0001
0.0000
car
0.0893
0.0399
0.0048
0.0069
0.8591
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.945
0.832
0.765
0.848
0.431
--
Recall/Completeness
0.810
0.944
0.767
0.919
0.859
--
F1
0.872
0.884
0.766
0.882
0.574
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8596
0.0324
0.0435
0.0458
0.0187
0.0000
building
0.1822
0.7832
0.0205
0.0100
0.0040
0.0000
low_veg
0.0692
0.0383
0.6385
0.2515
0.0025
0.0000
tree
0.0177
0.0104
0.0912
0.8801
0.0007
0.0000
car
0.1287
0.0000
0.0026
0.0084
0.8604
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.663
0.884
0.808
0.800
0.496
--
Recall/Completeness
0.860
0.783
0.638
0.880
0.860
--
F1
0.749
0.831
0.713
0.838
0.630
--

Overall accuracy 0.788

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
0.8905
0.0285
0.0365
0.0306
0.0139
0.0000
building
0.1888
0.7904
0.0118
0.0049
0.0042
0.0000
low_veg
0.0662
0.0256
0.6743
0.2313
0.0025
0.0000
tree
0.0119
0.0053
0.0777
0.9046
0.0006
0.0000
car
0.0552
0.0000
0.0013
0.0044
0.9392
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.679
0.920
0.840
0.826
0.482
--
Recall/Completeness
0.891
0.790
0.674
0.905
0.939
--
F1
0.771
0.850
0.748
0.863
0.637
--

Overall accuracy 0.815

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
0.8090
0.0721
0.0744
0.0329
0.0117
0.0000
building
0.0310
0.9252
0.0279
0.0135
0.0025
0.0000
low_veg
0.0909
0.0532
0.6954
0.1560
0.0045
0.0000
tree
0.0409
0.0117
0.1041
0.8425
0.0007
0.0000
car
0.0701
0.0011
0.0005
0.0106
0.9177
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.814
0.827
0.757
0.851
0.351
--
Recall/Completeness
0.809
0.925
0.695
0.843
0.918
--
F1
0.811
0.873
0.725
0.847
0.508
--

Overall accuracy 0.812

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
0.8290
0.0693
0.0691
0.0230
0.0096
0.0000
building
0.0225
0.9494
0.0175
0.0083
0.0022
0.0000
low_veg
0.0839
0.0383
0.7300
0.1431
0.0047
0.0000
tree
0.0320
0.0065
0.0928
0.8682
0.0006
0.0000
car
0.0261
0.0003
0.0000
0.0050
0.9686
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.844
0.856
0.782
0.874
0.324
--
Recall/Completeness
0.829
0.949
0.730
0.868
0.969
--
F1
0.837
0.900
0.755
0.871
0.486
--

Overall accuracy 0.839

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8170
0.0248
0.0826
0.0609
0.0147
0.0000
building
0.0428
0.9164
0.0265
0.0134
0.0010
0.0000
low_veg
0.0356
0.0218
0.6789
0.2620
0.0018
0.0000
tree
0.0291
0.0090
0.0796
0.8812
0.0010
0.0000
car
0.1004
0.0053
0.0054
0.0035
0.8854
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.813
0.923
0.812
0.783
0.372
--
Recall/Completeness
0.817
0.916
0.679
0.881
0.885
--
F1
0.815
0.920
0.739
0.829
0.524
--

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_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8504
0.0184
0.0770
0.0425
0.0117
0.0000
building
0.0363
0.9351
0.0201
0.0077
0.0009
0.0000
low_veg
0.0314
0.0123
0.7051
0.2493
0.0018
0.0000
tree
0.0206
0.0042
0.0707
0.9036
0.0009
0.0000
car
0.0517
0.0011
0.0028
0.0003
0.9442
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.845
0.955
0.836
0.802
0.348
--
Recall/Completeness
0.850
0.935
0.705
0.904
0.944
--
F1
0.848
0.945
0.765
0.850
0.509
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8280
0.0332
0.0852
0.0385
0.0151
0.0000
building
0.0160
0.9221
0.0389
0.0214
0.0017
0.0000
low_veg
0.0568
0.0232
0.7635
0.1546
0.0019
0.0000
tree
0.0148
0.0024
0.0674
0.9154
0.0001
0.0000
car
0.1366
0.0012
0.0150
0.0021
0.8451
0.0000
clutter
0.0174
0.8671
0.1114
0.0040
0.0000
0.0000
Precision/Correctness
0.889
0.809
0.732
0.901
0.483
-nan
Recall/Completeness
0.828
0.922
0.764
0.915
0.845
0.000
F1
0.858
0.862
0.747
0.908
0.615
-nan

Overall accuracy 0.844

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8619
0.0235
0.0781
0.0247
0.0117
0.0000
building
0.0137
0.9364
0.0315
0.0168
0.0016
0.0000
low_veg
0.0512
0.0119
0.8073
0.1276
0.0020
0.0000
tree
0.0090
0.0008
0.0529
0.9374
0.0000
0.0000
car
0.0869
0.0001
0.0037
0.0001
0.9092
0.0000
clutter
0.0125
0.8858
0.0994
0.0023
0.0000
0.0000
Precision/Correctness
0.915
0.829
0.765
0.926
0.472
-nan
Recall/Completeness
0.862
0.936
0.807
0.937
0.909
0.000
F1
0.888
0.879
0.786
0.931
0.622
-nan

Overall accuracy 0.871

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
0.8662
0.0247
0.0559
0.0334
0.0198
0.0000
building
0.0538
0.9040
0.0279
0.0118
0.0024
0.0000
low_veg
0.1086
0.0462
0.6650
0.1765
0.0037
0.0000
tree
0.0356
0.0055
0.0798
0.8785
0.0007
0.0000
car
0.2332
0.0005
0.0454
0.0098
0.7111
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.862
0.896
0.743
0.839
0.503
--
Recall/Completeness
0.866
0.904
0.665
0.879
0.711
--
F1
0.864
0.900
0.702
0.858
0.589
--

Overall accuracy 0.834

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
0.8987
0.0183
0.0476
0.0197
0.0158
0.0000
building
0.0438
0.9282
0.0184
0.0074
0.0022
0.0000
low_veg
0.1046
0.0324
0.7056
0.1537
0.0038
0.0000
tree
0.0221
0.0026
0.0629
0.9121
0.0003
0.0000
car
0.1802
0.0005
0.0354
0.0033
0.7807
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.890
0.928
0.788
0.874
0.500
--
Recall/Completeness
0.899
0.928
0.706
0.912
0.781
--
F1
0.894
0.928
0.744
0.893
0.610
--

Overall accuracy 0.869

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
0.7694
0.0398
0.1133
0.0553
0.0223
0.0000
building
0.0776
0.8579
0.0447
0.0157
0.0042
0.0000
low_veg
0.0254
0.0205
0.8127
0.1397
0.0018
0.0000
tree
0.0191
0.0100
0.1160
0.8541
0.0009
0.0000
car
0.1771
0.0000
0.0171
0.0002
0.8056
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.786
0.906
0.860
0.760
0.379
--
Recall/Completeness
0.769
0.858
0.813
0.854
0.806
--
F1
0.777
0.881
0.836
0.804
0.516
--

Overall accuracy 0.825

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
0.8066
0.0301
0.1027
0.0432
0.0175
0.0000
building
0.0738
0.8739
0.0365
0.0118
0.0040
0.0000
low_veg
0.0222
0.0154
0.8354
0.1253
0.0017
0.0000
tree
0.0131
0.0060
0.1010
0.8794
0.0006
0.0000
car
0.1282
0.0000
0.0082
0.0000
0.8637
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.810
0.932
0.882
0.783
0.361
--
Recall/Completeness
0.807
0.874
0.835
0.879
0.864
--
F1
0.809
0.902
0.858
0.828
0.510
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8822
0.0299
0.0318
0.0103
0.0457
0.0000
building
0.0643
0.8996
0.0197
0.0110
0.0054
0.0000
low_veg
0.0950
0.0299
0.6946
0.1745
0.0060
0.0000
tree
0.0131
0.0091
0.0672
0.9093
0.0013
0.0000
car
0.1723
0.0058
0.0113
0.0085
0.8021
0.0000
clutter
0.6127
0.0672
0.0940
0.0157
0.2104
0.0000
Precision/Correctness
0.851
0.950
0.848
0.687
0.588
-nan
Recall/Completeness
0.882
0.900
0.695
0.909
0.802
0.000
F1
0.866
0.924
0.764
0.782
0.679
-nan

Overall accuracy 0.849

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
0.9228
0.0193
0.0234
0.0052
0.0293
0.0000
building
0.0523
0.9219
0.0129
0.0085
0.0046
0.0000
low_veg
0.0800
0.0161
0.7388
0.1603
0.0047
0.0000
tree
0.0068
0.0054
0.0512
0.9362
0.0005
0.0000
car
0.1183
0.0030
0.0076
0.0067
0.8644
0.0000
clutter
0.5994
0.0600
0.0730
0.0126
0.2550
0.0000
Precision/Correctness
0.883
0.971
0.889
0.722
0.630
-nan
Recall/Completeness
0.923
0.922
0.739
0.936
0.864
0.000
F1
0.902
0.946
0.807
0.815
0.729
-nan

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9004
0.0206
0.0270
0.0213
0.0307
0.0000
building
0.0536
0.9088
0.0140
0.0117
0.0119
0.0000
low_veg
0.0656
0.0359
0.5689
0.3229
0.0066
0.0000
tree
0.0097
0.0031
0.0230
0.9642
0.0000
0.0000
car
0.1389
0.0077
0.0031
0.0111
0.8392
0.0000
clutter
0.4022
0.2377
0.0383
0.0130
0.3087
0.0000
Precision/Correctness
0.907
0.920
0.817
0.636
0.299
-nan
Recall/Completeness
0.900
0.909
0.569
0.964
0.839
0.000
F1
0.904
0.914
0.671
0.766
0.441
-nan

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9280
0.0132
0.0205
0.0129
0.0254
0.0000
building
0.0495
0.9219
0.0087
0.0078
0.0121
0.0000
low_veg
0.0465
0.0209
0.6158
0.3116
0.0052
0.0000
tree
0.0047
0.0012
0.0160
0.9781
0.0000
0.0000
car
0.0918
0.0042
0.0016
0.0079
0.8946
0.0000
clutter
0.3762
0.2429
0.0324
0.0067
0.3417
0.0000
Precision/Correctness
0.926
0.944
0.859
0.673
0.273
-nan
Recall/Completeness
0.928
0.922
0.616
0.978
0.895
0.000
F1
0.927
0.933
0.717
0.797
0.419
-nan

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9085
0.0137
0.0123
0.0121
0.0534
0.0000
building
0.2219
0.7489
0.0126
0.0033
0.0132
0.0000
low_veg
0.0589
0.0059
0.8857
0.0459
0.0036
0.0000
tree
0.0235
0.0047
0.0927
0.8748
0.0043
0.0000
car
0.1031
0.0556
0.0166
0.0076
0.8171
0.0000
clutter
0.3993
0.3131
0.0584
0.0239
0.2054
0.0000
Precision/Correctness
0.797
0.915
0.861
0.880
0.476
-nan
Recall/Completeness
0.909
0.749
0.886
0.875
0.817
0.000
F1
0.849
0.824
0.873
0.877
0.602
-nan

Overall accuracy 0.823

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
0.9369
0.0095
0.0087
0.0052
0.0397
0.0000
building
0.2181
0.7615
0.0071
0.0012
0.0121
0.0000
low_veg
0.0390
0.0032
0.9212
0.0349
0.0017
0.0000
tree
0.0101
0.0014
0.0782
0.9089
0.0014
0.0000
car
0.0677
0.0643
0.0093
0.0038
0.8550
0.0000
clutter
0.3578
0.3541
0.0576
0.0140
0.2165
0.0000
Precision/Correctness
0.815
0.929
0.896
0.925
0.486
-nan
Recall/Completeness
0.937
0.762
0.921
0.909
0.855
0.000
F1
0.872
0.837
0.908
0.917
0.620
-nan

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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9478
0.0112
0.0139
0.0043
0.0229
0.0000
building
0.2104
0.6940
0.0322
0.0550
0.0084
0.0000
low_veg
0.0783
0.0119
0.5009
0.4018
0.0072
0.0000
tree
0.0387
0.0093
0.1739
0.7750
0.0032
0.0000
car
0.2049
0.0035
0.0195
0.0005
0.7716
0.0000
clutter
0.4676
0.0940
0.0194
0.0023
0.4167
0.0000
Precision/Correctness
0.777
0.943
0.693
0.505
0.219
-nan
Recall/Completeness
0.948
0.694
0.501
0.775
0.772
0.000
F1
0.854
0.799
0.581
0.612
0.342
-nan

Overall accuracy 0.734

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
0.9668
0.0074
0.0081
0.0015
0.0163
0.0000
building
0.2070
0.7033
0.0262
0.0554
0.0081
0.0000
low_veg
0.0502
0.0072
0.5022
0.4352
0.0052
0.0000
tree
0.0186
0.0052
0.1604
0.8137
0.0021
0.0000
car
0.1601
0.0014
0.0153
0.0000
0.8233
0.0000
clutter
0.4302
0.0929
0.0177
0.0012
0.4580
0.0000
Precision/Correctness
0.797
0.959
0.728
0.501
0.201
-nan
Recall/Completeness
0.967
0.703
0.502
0.814
0.823
0.000
F1
0.874
0.811
0.594
0.620
0.323
-nan

Overall accuracy 0.752

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
0.8165
0.0368
0.0958
0.0359
0.0151
0.0000
building
0.0544
0.8854
0.0421
0.0089
0.0093
0.0000
low_veg
0.0431
0.0167
0.6962
0.2411
0.0029
0.0000
tree
0.0285
0.0057
0.1208
0.8442
0.0007
0.0000
car
0.1653
0.0147
0.0159
0.0179
0.7862
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.861
0.941
0.731
0.735
0.453
--
Recall/Completeness
0.817
0.885
0.696
0.844
0.786
--
F1
0.838
0.912
0.713
0.786
0.575
--

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_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8748
0.0313
0.0672
0.0158
0.0110
0.0000
building
0.0414
0.9205
0.0260
0.0036
0.0085
0.0000
low_veg
0.0338
0.0083
0.7240
0.2314
0.0024
0.0000
tree
0.0134
0.0023
0.0952
0.8889
0.0002
0.0000
car
0.0980
0.0183
0.0128
0.0122
0.8586
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.901
0.962
0.795
0.767
0.466
--
Recall/Completeness
0.875
0.920
0.724
0.889
0.859
--
F1
0.888
0.941
0.758
0.824
0.604
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8601
0.0306
0.0551
0.0254
0.0288
0.0000
building
0.0248
0.9038
0.0549
0.0097
0.0067
0.0000
low_veg
0.0445
0.0208
0.7914
0.1391
0.0043
0.0000
tree
0.0235
0.0045
0.1192
0.8517
0.0011
0.0000
car
0.1369
0.0206
0.0534
0.0202
0.7689
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.903
0.938
0.809
0.774
0.406
--
Recall/Completeness
0.860
0.904
0.791
0.852
0.769
--
F1
0.881
0.921
0.800
0.811
0.532
--

Overall accuracy 0.849

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
0.8976
0.0262
0.0395
0.0125
0.0242
0.0000
building
0.0168
0.9294
0.0421
0.0058
0.0060
0.0000
low_veg
0.0307
0.0101
0.8312
0.1245
0.0036
0.0000
tree
0.0131
0.0015
0.0975
0.8873
0.0006
0.0000
car
0.0816
0.0200
0.0442
0.0135
0.8407
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.938
0.960
0.850
0.810
0.400
--
Recall/Completeness
0.898
0.929
0.831
0.887
0.841
--
F1
0.917
0.944
0.840
0.847
0.542
--

Overall accuracy 0.885

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image