Detailled semantic labeling (2D) result


Name S. Ghassemi
Affiliation Politecnico di Torino, Department of Electronic and Telecommunication, Turin, Italy
Abbreviation TOR_2
Strategy (u)nsupervised, (s)upervised, (h)ybrid s


Overall statistics, reference set: full_reference


Number of non processed tiles by this participant: 0
↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.878
0.043
0.055
0.018
0.006
0.001
building
0.033
0.938
0.022
0.006
0.000
0.001
low_veg
0.067
0.025
0.779
0.127
0.001
0.001
tree
0.021
0.006
0.110
0.862
0.000
0.000
car
0.206
0.065
0.012
0.007
0.708
0.004
clutter
0.283
0.281
0.051
0.043
0.009
0.334
Precision/Correctness
0.881
0.917
0.784
0.850
0.831
0.749
Recall/Completeness
0.878
0.938
0.779
0.862
0.708
0.334
F1
0.880
0.928
0.781
0.856
0.764
0.462

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
0.906
0.036
0.046
0.008
0.003
0.001
building
0.025
0.956
0.015
0.003
0.000
0.001
low_veg
0.057
0.017
0.812
0.113
0.000
0.001
tree
0.011
0.003
0.094
0.892
0.000
0.000
car
0.118
0.074
0.008
0.004
0.792
0.004
clutter
0.247
0.300
0.046
0.040
0.008
0.360
Precision/Correctness
0.911
0.936
0.818
0.881
0.872
0.772
Recall/Completeness
0.906
0.956
0.812
0.892
0.792
0.360
F1
0.908
0.946
0.815
0.887
0.830
0.491

Overall accuracy 0.890
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9350
0.0160
0.0262
0.0152
0.0075
0.0002
building
0.0248
0.9522
0.0181
0.0041
0.0004
0.0004
low_veg
0.0622
0.0308
0.7496
0.1571
0.0002
0.0000
tree
0.0225
0.0070
0.0903
0.8780
0.0004
0.0017
car
0.2015
0.0017
0.0009
0.0006
0.7953
0.0000
clutter
0.0000
0.0000
0.0000
0.0120
0.0000
0.9880
Precision/Correctness
0.888
0.943
0.847
0.833
0.855
0.971
Recall/Completeness
0.935
0.952
0.750
0.878
0.795
0.988
F1
0.911
0.947
0.796
0.855
0.824
0.979

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9624
0.0125
0.0148
0.0052
0.0049
0.0002
building
0.0185
0.9712
0.0082
0.0014
0.0003
0.0004
low_veg
0.0464
0.0232
0.7808
0.1494
0.0002
0.0000
tree
0.0097
0.0044
0.0771
0.9077
0.0002
0.0008
car
0.0982
0.0005
0.0001
0.0001
0.9010
0.0000
clutter
0.0000
0.0000
0.0000
0.0029
0.0000
0.9971
Precision/Correctness
0.925
0.959
0.883
0.858
0.886
0.983
Recall/Completeness
0.962
0.971
0.781
0.908
0.901
0.997
F1
0.944
0.965
0.829
0.882
0.894
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
0.7926
0.0911
0.1052
0.0096
0.0015
0.0000
building
0.0532
0.9184
0.0226
0.0058
0.0000
0.0000
low_veg
0.0710
0.0297
0.5720
0.3268
0.0005
0.0000
tree
0.0294
0.0061
0.0813
0.8831
0.0000
0.0000
car
0.4079
0.0092
0.0502
0.0061
0.5265
0.0001
clutter
--
--
--
--
--
--
Precision/Correctness
0.851
0.948
0.347
0.671
0.942
--
Recall/Completeness
0.793
0.918
0.572
0.883
0.527
--
F1
0.821
0.933
0.432
0.763
0.676
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8190
0.0727
0.1026
0.0052
0.0004
0.0000
building
0.0435
0.9335
0.0193
0.0038
0.0000
0.0000
low_veg
0.0393
0.0080
0.6030
0.3493
0.0004
0.0000
tree
0.0098
0.0023
0.0683
0.9196
0.0000
0.0000
car
0.3042
0.0033
0.0446
0.0041
0.6436
0.0001
clutter
--
--
--
--
--
--
Precision/Correctness
0.883
0.963
0.353
0.709
0.981
--
Recall/Completeness
0.819
0.933
0.603
0.920
0.644
--
F1
0.850
0.948
0.445
0.801
0.777
--

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_area6_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9175
0.0337
0.0324
0.0110
0.0053
0.0000
building
0.0474
0.9322
0.0171
0.0030
0.0003
0.0000
low_veg
0.0877
0.0176
0.7616
0.1322
0.0008
0.0000
tree
0.0146
0.0030
0.0741
0.9080
0.0003
0.0000
car
0.2710
0.0176
0.0018
0.0005
0.7088
0.0003
clutter
--
--
--
--
--
--
Precision/Correctness
0.874
0.964
0.697
0.936
0.876
--
Recall/Completeness
0.918
0.932
0.762
0.908
0.709
--
F1
0.895
0.948
0.728
0.922
0.784
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9498
0.0216
0.0233
0.0036
0.0017
0.0000
building
0.0361
0.9521
0.0108
0.0008
0.0002
0.0000
low_veg
0.0718
0.0067
0.8075
0.1139
0.0002
0.0000
tree
0.0057
0.0007
0.0596
0.9339
0.0001
0.0000
car
0.1899
0.0164
0.0007
0.0000
0.7926
0.0004
clutter
--
--
--
--
--
--
Precision/Correctness
0.910
0.981
0.743
0.958
0.947
--
Recall/Completeness
0.950
0.952
0.808
0.934
0.793
--
F1
0.929
0.966
0.774
0.946
0.863
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8697
0.0337
0.0712
0.0192
0.0053
0.0008
building
0.0266
0.9549
0.0147
0.0036
0.0001
0.0000
low_veg
0.0455
0.0145
0.8714
0.0685
0.0001
0.0000
tree
0.0178
0.0062
0.0676
0.9078
0.0001
0.0004
car
0.1791
0.0032
0.0228
0.0100
0.7850
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.924
0.936
0.802
0.914
0.885
--
Recall/Completeness
0.870
0.955
0.871
0.908
0.785
--
F1
0.896
0.946
0.835
0.911
0.832
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9010
0.0271
0.0609
0.0079
0.0023
0.0007
building
0.0176
0.9744
0.0066
0.0013
0.0001
0.0000
low_veg
0.0325
0.0073
0.9045
0.0557
0.0000
0.0000
tree
0.0066
0.0024
0.0495
0.9413
0.0000
0.0003
car
0.0946
0.0015
0.0139
0.0082
0.8818
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.955
0.957
0.843
0.943
0.937
--
Recall/Completeness
0.901
0.974
0.904
0.941
0.882
--
F1
0.927
0.966
0.873
0.942
0.909
--

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_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7744
0.1422
0.0556
0.0208
0.0069
0.0000
building
0.0206
0.9489
0.0149
0.0128
0.0015
0.0012
low_veg
0.0872
0.0283
0.7446
0.1399
0.0000
0.0000
tree
0.0135
0.0073
0.0897
0.8895
0.0000
0.0001
car
0.1654
0.0233
0.0043
0.0022
0.8032
0.0015
clutter
--
--
--
--
--
--
Precision/Correctness
0.923
0.789
0.753
0.838
0.658
--
Recall/Completeness
0.774
0.949
0.745
0.889
0.803
--
F1
0.842
0.861
0.749
0.863
0.723
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8013
0.1359
0.0482
0.0091
0.0055
0.0000
building
0.0149
0.9622
0.0106
0.0098
0.0013
0.0012
low_veg
0.0850
0.0163
0.7902
0.1085
0.0000
0.0000
tree
0.0056
0.0033
0.0716
0.9195
0.0000
0.0000
car
0.0906
0.0263
0.0012
0.0000
0.8811
0.0008
clutter
--
--
--
--
--
--
Precision/Correctness
0.940
0.806
0.791
0.884
0.654
--
Recall/Completeness
0.801
0.962
0.790
0.919
0.881
--
F1
0.865
0.877
0.790
0.901
0.751
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8814
0.0505
0.0420
0.0194
0.0057
0.0009
building
0.0238
0.9507
0.0064
0.0038
0.0005
0.0149
low_veg
0.1009
0.0298
0.7629
0.1021
0.0008
0.0035
tree
0.0173
0.0097
0.1432
0.8294
0.0001
0.0003
car
0.0892
0.0000
0.0000
0.0006
0.9102
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.755
0.903
0.786
0.903
0.796
--
Recall/Completeness
0.881
0.951
0.763
0.829
0.910
--
F1
0.813
0.926
0.774
0.864
0.850
--

Overall accuracy 0.841

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
0.9066
0.0486
0.0326
0.0094
0.0022
0.0006
building
0.0212
0.9593
0.0018
0.0012
0.0005
0.0160
low_veg
0.0993
0.0198
0.7952
0.0812
0.0007
0.0038
tree
0.0106
0.0034
0.1256
0.8601
0.0000
0.0002
car
0.0201
0.0000
0.0000
0.0000
0.9799
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.781
0.931
0.813
0.929
0.853
--
Recall/Completeness
0.907
0.959
0.795
0.860
0.980
--
F1
0.839
0.945
0.804
0.893
0.912
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8499
0.0379
0.0728
0.0314
0.0078
0.0001
building
0.0151
0.9636
0.0159
0.0032
0.0003
0.0019
low_veg
0.0874
0.0532
0.7165
0.1410
0.0016
0.0003
tree
0.0298
0.0135
0.1456
0.8108
0.0001
0.0002
car
0.1002
0.0000
0.0000
0.0025
0.8973
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.847
0.864
0.727
0.862
0.512
--
Recall/Completeness
0.850
0.964
0.716
0.811
0.897
--
F1
0.849
0.911
0.722
0.835
0.652
--

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_area14_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8702
0.0344
0.0683
0.0202
0.0068
0.0000
building
0.0099
0.9793
0.0082
0.0007
0.0001
0.0018
low_veg
0.0823
0.0384
0.7458
0.1316
0.0016
0.0003
tree
0.0221
0.0063
0.1344
0.8372
0.0000
0.0001
car
0.0234
0.0000
0.0000
0.0000
0.9766
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.872
0.897
0.747
0.883
0.488
--
Recall/Completeness
0.870
0.979
0.746
0.837
0.977
--
F1
0.871
0.936
0.746
0.859
0.651
--

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
0.8362
0.0619
0.0579
0.0379
0.0061
0.0002
building
0.0120
0.9623
0.0130
0.0067
0.0001
0.0058
low_veg
0.0663
0.0218
0.8277
0.0802
0.0010
0.0030
tree
0.0294
0.0077
0.1673
0.7951
0.0001
0.0003
car
0.0897
0.0008
0.0005
0.0000
0.9090
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.798
0.901
0.772
0.904
0.611
--
Recall/Completeness
0.836
0.962
0.828
0.795
0.909
--
F1
0.816
0.930
0.799
0.846
0.731
--

Overall accuracy 0.841

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
0.8670
0.0573
0.0488
0.0229
0.0037
0.0002
building
0.0083
0.9795
0.0046
0.0020
0.0000
0.0056
low_veg
0.0638
0.0142
0.8504
0.0679
0.0009
0.0029
tree
0.0206
0.0035
0.1541
0.8215
0.0001
0.0002
car
0.0338
0.0000
0.0000
0.0000
0.9662
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.823
0.925
0.796
0.926
0.639
--
Recall/Completeness
0.867
0.980
0.850
0.821
0.966
--
F1
0.845
0.952
0.822
0.871
0.769
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8307
0.0839
0.0325
0.0345
0.0088
0.0096
building
0.0193
0.9705
0.0044
0.0040
0.0009
0.0009
low_veg
0.1317
0.0259
0.7161
0.1241
0.0021
0.0002
tree
0.0123
0.0028
0.0722
0.9118
0.0001
0.0009
car
0.0690
0.0001
0.0035
0.0016
0.9259
0.0000
clutter
0.0724
0.9260
0.0005
0.0001
0.0000
0.0011
Precision/Correctness
0.823
0.767
0.790
0.922
0.619
0.012
Recall/Completeness
0.831
0.971
0.716
0.912
0.926
0.001
F1
0.827
0.857
0.751
0.917
0.742
0.002

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8602
0.0765
0.0287
0.0192
0.0058
0.0095
building
0.0162
0.9800
0.0014
0.0008
0.0010
0.0006
low_veg
0.1265
0.0135
0.7592
0.0985
0.0022
0.0001
tree
0.0059
0.0008
0.0588
0.9339
0.0000
0.0006
car
0.0194
0.0000
0.0000
0.0000
0.9806
0.0000
clutter
0.0638
0.9357
0.0000
0.0000
0.0000
0.0005
Precision/Correctness
0.851
0.785
0.821
0.946
0.625
0.006
Recall/Completeness
0.860
0.980
0.759
0.934
0.981
0.000
F1
0.855
0.872
0.789
0.940
0.764
0.001

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8923
0.0254
0.0416
0.0313
0.0095
0.0000
building
0.0219
0.9604
0.0128
0.0046
0.0002
0.0000
low_veg
0.1120
0.0427
0.7141
0.1302
0.0009
0.0001
tree
0.0221
0.0039
0.0776
0.8961
0.0003
0.0000
car
0.1366
0.0006
0.0055
0.0138
0.8429
0.0006
clutter
--
--
--
--
--
--
Precision/Correctness
0.892
0.905
0.791
0.874
0.735
--
Recall/Completeness
0.892
0.960
0.714
0.896
0.843
--
F1
0.892
0.932
0.751
0.885
0.785
--

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_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9251
0.0185
0.0347
0.0155
0.0062
0.0000
building
0.0167
0.9756
0.0062
0.0014
0.0001
0.0000
low_veg
0.1084
0.0312
0.7568
0.1029
0.0006
0.0002
tree
0.0095
0.0009
0.0605
0.9291
0.0001
0.0000
car
0.0612
0.0000
0.0027
0.0061
0.9296
0.0003
clutter
--
--
--
--
--
--
Precision/Correctness
0.916
0.935
0.832
0.913
0.778
--
Recall/Completeness
0.925
0.976
0.757
0.929
0.930
--
F1
0.921
0.955
0.793
0.921
0.847
--

Overall accuracy 0.903

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
0.8202
0.0489
0.0823
0.0228
0.0144
0.0115
building
0.0433
0.9308
0.0168
0.0091
0.0000
0.0000
low_veg
0.0406
0.0206
0.8843
0.0503
0.0006
0.0035
tree
0.0254
0.0112
0.1925
0.7700
0.0001
0.0008
car
0.0879
0.0000
0.0003
0.0063
0.9054
0.0001
clutter
--
--
--
--
--
--
Precision/Correctness
0.790
0.905
0.850
0.884
0.590
--
Recall/Completeness
0.820
0.931
0.884
0.770
0.905
--
F1
0.805
0.918
0.867
0.823
0.715
--

Overall accuracy 0.855

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8523
0.0365
0.0740
0.0148
0.0110
0.0113
building
0.0422
0.9389
0.0118
0.0071
0.0000
0.0000
low_veg
0.0366
0.0145
0.9039
0.0408
0.0006
0.0036
tree
0.0185
0.0065
0.1713
0.8031
0.0001
0.0006
car
0.0338
0.0000
0.0000
0.0012
0.9649
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.813
0.933
0.872
0.907
0.585
--
Recall/Completeness
0.852
0.939
0.904
0.803
0.965
--
F1
0.832
0.936
0.887
0.852
0.729
--

Overall accuracy 0.878

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
0.9241
0.0175
0.0417
0.0096
0.0071
0.0000
building
0.0658
0.9033
0.0244
0.0065
0.0000
0.0000
low_veg
0.0543
0.0192
0.7036
0.2228
0.0002
0.0000
tree
0.0088
0.0020
0.0353
0.9538
0.0002
0.0000
car
0.2333
0.0026
0.0111
0.0070
0.7460
0.0000
clutter
0.6823
0.0198
0.2105
0.0057
0.0817
0.0000
Precision/Correctness
0.871
0.970
0.840
0.660
0.906
-nan
Recall/Completeness
0.924
0.903
0.704
0.954
0.746
0.000
F1
0.897
0.936
0.766
0.780
0.818
-nan

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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9516
0.0098
0.0328
0.0039
0.0019
0.0000
building
0.0544
0.9258
0.0173
0.0025
0.0000
0.0000
low_veg
0.0391
0.0088
0.7426
0.2095
0.0000
0.0000
tree
0.0032
0.0002
0.0218
0.9748
0.0000
0.0000
car
0.1287
0.0022
0.0082
0.0053
0.8556
0.0000
clutter
0.7144
0.0116
0.1943
0.0025
0.0773
0.0000
Precision/Correctness
0.904
0.986
0.878
0.694
0.965
-nan
Recall/Completeness
0.952
0.926
0.743
0.975
0.856
0.000
F1
0.927
0.955
0.805
0.811
0.907
-nan

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9094
0.0186
0.0497
0.0172
0.0048
0.0002
building
0.0261
0.9344
0.0306
0.0087
0.0002
0.0000
low_veg
0.0731
0.0129
0.6211
0.2927
0.0002
0.0000
tree
0.0024
0.0009
0.0234
0.9733
0.0000
0.0000
car
0.2210
0.0018
0.0005
0.0138
0.7395
0.0234
clutter
0.4795
0.3495
0.0941
0.0191
0.0082
0.0496
Precision/Correctness
0.916
0.926
0.726
0.665
0.808
0.794
Recall/Completeness
0.909
0.934
0.621
0.973
0.740
0.050
F1
0.913
0.930
0.669
0.790
0.772
0.093

Overall accuracy 0.853

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9348
0.0139
0.0395
0.0084
0.0034
0.0000
building
0.0215
0.9519
0.0217
0.0047
0.0002
0.0000
low_veg
0.0618
0.0051
0.6551
0.2779
0.0001
0.0000
tree
0.0004
0.0000
0.0152
0.9844
0.0000
0.0000
car
0.1332
0.0000
0.0000
0.0104
0.8297
0.0267
clutter
0.4646
0.3719
0.0899
0.0118
0.0078
0.0540
Precision/Correctness
0.934
0.941
0.768
0.707
0.831
0.833
Recall/Completeness
0.935
0.952
0.655
0.984
0.830
0.054
F1
0.934
0.946
0.707
0.823
0.831
0.101

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
0.8824
0.0226
0.0816
0.0099
0.0028
0.0008
building
0.0252
0.9445
0.0250
0.0050
0.0001
0.0002
low_veg
0.0201
0.0024
0.9289
0.0484
0.0000
0.0002
tree
0.0231
0.0022
0.0951
0.8793
0.0003
0.0000
car
0.1634
0.3496
0.0172
0.0119
0.4454
0.0125
clutter
0.3472
0.2561
0.1179
0.1388
0.0191
0.1209
Precision/Correctness
0.919
0.895
0.719
0.851
0.900
0.831
Recall/Completeness
0.882
0.944
0.929
0.879
0.445
0.121
F1
0.900
0.919
0.811
0.865
0.596
0.211

Overall accuracy 0.865

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
0.9018
0.0159
0.0775
0.0033
0.0011
0.0005
building
0.0164
0.9642
0.0176
0.0017
0.0000
0.0001
low_veg
0.0092
0.0010
0.9519
0.0379
0.0000
0.0000
tree
0.0090
0.0005
0.0774
0.9131
0.0000
0.0000
car
0.0884
0.3857
0.0088
0.0078
0.4950
0.0143
clutter
0.2967
0.2882
0.1130
0.1480
0.0189
0.1352
Precision/Correctness
0.950
0.914
0.750
0.892
0.933
0.868
Recall/Completeness
0.902
0.964
0.952
0.913
0.495
0.135
F1
0.925
0.939
0.839
0.902
0.647
0.234

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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9523
0.0167
0.0203
0.0058
0.0013
0.0036
building
0.0235
0.9152
0.0534
0.0056
0.0000
0.0024
low_veg
0.0709
0.0153
0.7708
0.1431
0.0000
0.0000
tree
0.0309
0.0044
0.1126
0.8521
0.0000
0.0000
car
0.2810
0.0761
0.0086
0.0036
0.6287
0.0020
clutter
0.3463
0.0879
0.0141
0.0272
0.0048
0.5196
Precision/Correctness
0.894
0.947
0.767
0.765
0.914
0.930
Recall/Completeness
0.952
0.915
0.771
0.852
0.629
0.520
F1
0.922
0.931
0.769
0.806
0.745
0.667

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_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9690
0.0121
0.0153
0.0020
0.0005
0.0011
building
0.0173
0.9263
0.0517
0.0027
0.0000
0.0020
low_veg
0.0449
0.0088
0.8005
0.1458
0.0000
0.0000
tree
0.0132
0.0013
0.1044
0.8811
0.0000
0.0000
car
0.1827
0.0874
0.0077
0.0012
0.7182
0.0028
clutter
0.3073
0.0890
0.0136
0.0274
0.0041
0.5585
Precision/Correctness
0.924
0.961
0.786
0.783
0.947
0.963
Recall/Completeness
0.969
0.926
0.801
0.881
0.718
0.559
F1
0.946
0.943
0.793
0.829
0.817
0.707

Overall accuracy 0.899

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8508
0.0389
0.0864
0.0218
0.0019
0.0002
building
0.0190
0.9521
0.0233
0.0049
0.0002
0.0005
low_veg
0.0399
0.0124
0.7664
0.1811
0.0001
0.0001
tree
0.0284
0.0059
0.1262
0.8393
0.0001
0.0000
car
0.1640
0.0214
0.0032
0.0094
0.8017
0.0003
clutter
--
--
--
--
--
--
Precision/Correctness
0.903
0.946
0.768
0.792
0.913
--
Recall/Completeness
0.851
0.952
0.766
0.839
0.802
--
F1
0.876
0.949
0.767
0.815
0.854
--

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_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8969
0.0346
0.0611
0.0069
0.0003
0.0002
building
0.0118
0.9759
0.0110
0.0007
0.0001
0.0005
low_veg
0.0287
0.0056
0.7935
0.1721
0.0000
0.0001
tree
0.0103
0.0016
0.0990
0.8891
0.0000
0.0000
car
0.0776
0.0274
0.0033
0.0017
0.8899
0.0001
clutter
--
--
--
--
--
--
Precision/Correctness
0.944
0.964
0.827
0.822
0.976
--
Recall/Completeness
0.897
0.976
0.793
0.889
0.890
--
F1
0.920
0.970
0.810
0.854
0.931
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9024
0.0320
0.0443
0.0150
0.0043
0.0021
building
0.0222
0.9376
0.0338
0.0053
0.0011
0.0000
low_veg
0.0528
0.0287
0.8164
0.1017
0.0003
0.0001
tree
0.0214
0.0027
0.1285
0.8473
0.0001
0.0000
car
0.1712
0.0178
0.0106
0.0053
0.7950
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.902
0.932
0.834
0.830
0.831
--
Recall/Completeness
0.902
0.938
0.816
0.847
0.795
--
F1
0.902
0.935
0.825
0.839
0.813
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9293
0.0302
0.0307
0.0052
0.0024
0.0023
building
0.0145
0.9646
0.0179
0.0020
0.0009
0.0000
low_veg
0.0395
0.0253
0.8450
0.0899
0.0001
0.0001
tree
0.0084
0.0005
0.1044
0.8868
0.0000
0.0000
car
0.0902
0.0178
0.0026
0.0019
0.8875
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.936
0.943
0.877
0.863
0.879
--
Recall/Completeness
0.929
0.965
0.845
0.887
0.888
--
F1
0.933
0.954
0.861
0.875
0.883
--

Overall accuracy 0.907

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image