Detailled semantic labeling (2D) result


Name A. Lagrange and B. Le Saux (NiN LeNet MS MD)
Affiliation Onera, France,
Abbreviation ONE_4
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
0.851
0.047
0.052
0.020
0.029
0.001
building
0.102
0.851
0.033
0.008
0.004
0.001
low_veg
0.062
0.032
0.757
0.145
0.004
0.000
tree
0.023
0.008
0.129
0.838
0.001
0.000
car
0.192
0.061
0.015
0.006
0.725
0.002
clutter
0.224
0.459
0.017
0.009
0.139
0.153
Precision/Correctness
0.827
0.893
0.757
0.828
0.451
0.719
Recall/Completeness
0.851
0.851
0.757
0.838
0.725
0.153
F1
0.839
0.872
0.757
0.833
0.556
0.252

Overall accuracy 0.820

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
0.884
0.039
0.044
0.011
0.022
0.001
building
0.096
0.869
0.026
0.005
0.004
0.001
low_veg
0.051
0.024
0.791
0.131
0.003
0.000
tree
0.013
0.005
0.112
0.869
0.001
0.000
car
0.151
0.066
0.011
0.003
0.766
0.002
clutter
0.196
0.485
0.013
0.003
0.134
0.169
Precision/Correctness
0.852
0.915
0.790
0.859
0.456
0.756
Recall/Completeness
0.884
0.869
0.791
0.869
0.766
0.169
F1
0.868
0.891
0.790
0.864
0.571
0.276

Overall accuracy 0.850
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8932
0.0195
0.0402
0.0175
0.0286
0.0009
building
0.0635
0.8912
0.0353
0.0052
0.0045
0.0002
low_veg
0.0690
0.0234
0.7462
0.1563
0.0039
0.0012
tree
0.0247
0.0072
0.1276
0.8368
0.0017
0.0021
car
0.2311
0.0146
0.0024
0.0038
0.7475
0.0006
clutter
0.0557
0.0038
0.0000
0.0119
0.0000
0.9285
Precision/Correctness
0.841
0.942
0.786
0.824
0.549
0.940
Recall/Completeness
0.893
0.891
0.746
0.837
0.747
0.929
F1
0.867
0.916
0.766
0.830
0.633
0.934

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
0.9292
0.0148
0.0279
0.0066
0.0209
0.0006
building
0.0572
0.9124
0.0242
0.0021
0.0040
0.0001
low_veg
0.0553
0.0150
0.7774
0.1485
0.0030
0.0009
tree
0.0118
0.0042
0.1141
0.8679
0.0008
0.0011
car
0.1799
0.0129
0.0018
0.0017
0.8034
0.0003
clutter
0.0572
0.0035
0.0000
0.0027
0.0000
0.9366
Precision/Correctness
0.874
0.962
0.819
0.851
0.578
0.965
Recall/Completeness
0.929
0.912
0.777
0.868
0.803
0.937
F1
0.901
0.937
0.798
0.860
0.672
0.951

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
0.8738
0.0695
0.0166
0.0055
0.0339
0.0007
building
0.0824
0.9010
0.0104
0.0023
0.0034
0.0004
low_veg
0.1529
0.0603
0.4977
0.2688
0.0143
0.0061
tree
0.0553
0.0215
0.1275
0.7916
0.0042
0.0000
car
0.2446
0.0129
0.0056
0.0006
0.7363
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.811
0.954
0.568
0.718
0.458
--
Recall/Completeness
0.874
0.901
0.498
0.792
0.736
--
F1
0.841
0.927
0.530
0.753
0.565
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9058
0.0504
0.0131
0.0024
0.0277
0.0007
building
0.0680
0.9188
0.0086
0.0013
0.0030
0.0003
low_veg
0.1169
0.0354
0.5342
0.2974
0.0103
0.0058
tree
0.0282
0.0124
0.1239
0.8344
0.0012
0.0000
car
0.2129
0.0067
0.0042
0.0001
0.7760
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.842
0.970
0.594
0.746
0.458
--
Recall/Completeness
0.906
0.919
0.534
0.834
0.776
--
F1
0.873
0.944
0.562
0.788
0.576
--

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_area6_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8359
0.0627
0.0514
0.0145
0.0353
0.0003
building
0.0642
0.8996
0.0268
0.0050
0.0045
0.0000
low_veg
0.0741
0.0385
0.7554
0.1277
0.0043
0.0000
tree
0.0161
0.0064
0.0935
0.8834
0.0006
0.0001
car
0.2453
0.0430
0.0071
0.0031
0.6968
0.0048
clutter
--
--
--
--
--
--
Precision/Correctness
0.849
0.929
0.622
0.930
0.507
--
Recall/Completeness
0.836
0.900
0.755
0.883
0.697
--
F1
0.843
0.914
0.682
0.906
0.587
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8778
0.0475
0.0415
0.0062
0.0269
0.0001
building
0.0537
0.9198
0.0206
0.0025
0.0035
0.0000
low_veg
0.0588
0.0286
0.8004
0.1096
0.0025
0.0000
tree
0.0074
0.0028
0.0784
0.9111
0.0002
0.0000
car
0.2132
0.0312
0.0048
0.0014
0.7443
0.0052
clutter
--
--
--
--
--
--
Precision/Correctness
0.883
0.952
0.656
0.954
0.527
--
Recall/Completeness
0.878
0.920
0.800
0.911
0.744
--
F1
0.880
0.936
0.721
0.932
0.617
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8317
0.0425
0.0792
0.0158
0.0309
0.0000
building
0.0544
0.9082
0.0273
0.0071
0.0030
0.0000
low_veg
0.0622
0.0165
0.8371
0.0807
0.0035
0.0000
tree
0.0247
0.0074
0.1281
0.8387
0.0012
0.0000
car
0.2111
0.0237
0.0275
0.0068
0.7308
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.884
0.918
0.726
0.899
0.521
--
Recall/Completeness
0.832
0.908
0.837
0.839
0.731
--
F1
0.857
0.913
0.777
0.868
0.609
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8645
0.0373
0.0682
0.0067
0.0233
0.0000
building
0.0433
0.9326
0.0177
0.0040
0.0025
0.0000
low_veg
0.0464
0.0094
0.8708
0.0714
0.0020
0.0000
tree
0.0120
0.0039
0.1092
0.8744
0.0005
0.0000
car
0.1713
0.0244
0.0211
0.0044
0.7787
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.917
0.938
0.765
0.925
0.543
--
Recall/Completeness
0.865
0.933
0.871
0.874
0.779
--
F1
0.890
0.935
0.814
0.899
0.640
--

Overall accuracy 0.883

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7963
0.0682
0.0991
0.0208
0.0151
0.0006
building
0.0328
0.9216
0.0286
0.0135
0.0029
0.0007
low_veg
0.0684
0.0423
0.7358
0.1502
0.0033
0.0000
tree
0.0192
0.0076
0.1199
0.8522
0.0011
0.0000
car
0.1823
0.0364
0.0254
0.0096
0.7464
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.923
0.862
0.653
0.824
0.424
--
Recall/Completeness
0.796
0.922
0.736
0.852
0.746
--
F1
0.855
0.891
0.692
0.838
0.541
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8269
0.0587
0.0915
0.0100
0.0123
0.0006
building
0.0252
0.9404
0.0213
0.0102
0.0022
0.0007
low_veg
0.0639
0.0335
0.7717
0.1278
0.0031
0.0000
tree
0.0099
0.0043
0.0980
0.8870
0.0009
0.0000
car
0.1386
0.0401
0.0171
0.0036
0.8006
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.942
0.886
0.685
0.864
0.415
--
Recall/Completeness
0.827
0.940
0.772
0.887
0.801
--
F1
0.881
0.912
0.726
0.875
0.546
--

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
0.8598
0.0376
0.0420
0.0369
0.0236
0.0002
building
0.1522
0.8240
0.0140
0.0077
0.0021
0.0000
low_veg
0.0688
0.0515
0.7162
0.1611
0.0021
0.0004
tree
0.0179
0.0105
0.1550
0.8160
0.0006
0.0000
car
0.1041
0.0027
0.0131
0.0016
0.8785
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.685
0.867
0.759
0.850
0.492
--
Recall/Completeness
0.860
0.824
0.716
0.816
0.878
--
F1
0.763
0.845
0.737
0.833
0.630
--

Overall accuracy 0.795

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
0.8918
0.0342
0.0344
0.0218
0.0175
0.0002
building
0.1575
0.8326
0.0053
0.0026
0.0021
0.0000
low_veg
0.0654
0.0436
0.7488
0.1397
0.0020
0.0004
tree
0.0115
0.0064
0.1394
0.8423
0.0004
0.0000
car
0.0610
0.0019
0.0130
0.0004
0.9237
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.704
0.892
0.786
0.879
0.485
--
Recall/Completeness
0.892
0.833
0.749
0.842
0.924
--
F1
0.787
0.861
0.767
0.860
0.636
--

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7991
0.0381
0.1092
0.0394
0.0137
0.0005
building
0.0337
0.9245
0.0287
0.0122
0.0009
0.0000
low_veg
0.0981
0.0501
0.6807
0.1671
0.0035
0.0005
tree
0.0386
0.0137
0.1374
0.8099
0.0003
0.0001
car
0.1336
0.0000
0.0030
0.0046
0.8588
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.806
0.862
0.695
0.835
0.348
--
Recall/Completeness
0.799
0.925
0.681
0.810
0.859
--
F1
0.803
0.892
0.688
0.822
0.495
--

Overall accuracy 0.795

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
0.8181
0.0350
0.1062
0.0286
0.0117
0.0005
building
0.0235
0.9506
0.0185
0.0068
0.0006
0.0000
low_veg
0.0910
0.0391
0.7100
0.1558
0.0035
0.0006
tree
0.0294
0.0084
0.1278
0.8341
0.0003
0.0001
car
0.0824
0.0000
0.0004
0.0007
0.9165
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.838
0.889
0.713
0.858
0.323
--
Recall/Completeness
0.818
0.951
0.710
0.834
0.916
--
F1
0.828
0.919
0.711
0.846
0.477
--

Overall accuracy 0.820

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
0.7953
0.0409
0.0872
0.0625
0.0140
0.0000
building
0.0605
0.8871
0.0313
0.0191
0.0021
0.0000
low_veg
0.0307
0.0253
0.7885
0.1541
0.0014
0.0001
tree
0.0276
0.0090
0.1279
0.8346
0.0008
0.0000
car
0.1222
0.0006
0.0416
0.0139
0.8217
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.802
0.902
0.780
0.839
0.362
--
Recall/Completeness
0.795
0.887
0.788
0.835
0.822
--
F1
0.799
0.894
0.784
0.837
0.503
--

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_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8300
0.0366
0.0798
0.0421
0.0115
0.0000
building
0.0559
0.9116
0.0209
0.0099
0.0017
0.0000
low_veg
0.0255
0.0202
0.8128
0.1402
0.0012
0.0001
tree
0.0201
0.0054
0.1161
0.8577
0.0007
0.0000
car
0.0770
0.0000
0.0347
0.0030
0.8852
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.833
0.924
0.805
0.864
0.348
--
Recall/Completeness
0.830
0.912
0.813
0.858
0.885
--
F1
0.831
0.918
0.809
0.861
0.500
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8113
0.0559
0.0756
0.0443
0.0122
0.0007
building
0.0178
0.9180
0.0528
0.0103
0.0009
0.0001
low_veg
0.0542
0.0248
0.7475
0.1696
0.0025
0.0014
tree
0.0092
0.0039
0.0714
0.9153
0.0001
0.0001
car
0.2141
0.0194
0.0193
0.0190
0.7198
0.0083
clutter
0.0284
0.9591
0.0038
0.0052
0.0011
0.0025
Precision/Correctness
0.896
0.775
0.731
0.896
0.485
0.122
Recall/Completeness
0.811
0.918
0.748
0.915
0.720
0.002
F1
0.852
0.840
0.739
0.906
0.579
0.005

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8433
0.0489
0.0702
0.0272
0.0099
0.0005
building
0.0144
0.9344
0.0465
0.0038
0.0008
0.0001
low_veg
0.0490
0.0154
0.7888
0.1429
0.0025
0.0013
tree
0.0043
0.0023
0.0580
0.9352
0.0000
0.0000
car
0.1495
0.0217
0.0095
0.0118
0.7994
0.0081
clutter
0.0229
0.9726
0.0019
0.0008
0.0007
0.0011
Precision/Correctness
0.922
0.791
0.759
0.923
0.474
0.074
Recall/Completeness
0.843
0.934
0.789
0.935
0.799
0.001
F1
0.881
0.857
0.774
0.929
0.595
0.002

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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8567
0.0317
0.0526
0.0387
0.0199
0.0003
building
0.0486
0.9180
0.0228
0.0085
0.0021
0.0000
low_veg
0.0953
0.0452
0.7044
0.1513
0.0032
0.0006
tree
0.0238
0.0056
0.0972
0.8729
0.0006
0.0000
car
0.2248
0.0260
0.0512
0.0126
0.6854
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.880
0.886
0.742
0.850
0.498
--
Recall/Completeness
0.857
0.918
0.704
0.873
0.685
--
F1
0.868
0.901
0.723
0.861
0.577
--

Overall accuracy 0.840

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
0.8920
0.0250
0.0446
0.0225
0.0157
0.0002
building
0.0407
0.9412
0.0132
0.0033
0.0016
0.0000
low_veg
0.0918
0.0357
0.7425
0.1266
0.0028
0.0006
tree
0.0119
0.0028
0.0801
0.9050
0.0002
0.0000
car
0.1827
0.0317
0.0427
0.0056
0.7373
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.905
0.914
0.783
0.888
0.500
--
Recall/Completeness
0.892
0.941
0.743
0.905
0.737
--
F1
0.899
0.927
0.762
0.897
0.596
--

Overall accuracy 0.874

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
0.7605
0.0454
0.1296
0.0425
0.0221
0.0000
building
0.0829
0.8268
0.0719
0.0173
0.0010
0.0000
low_veg
0.0289
0.0201
0.8564
0.0936
0.0010
0.0000
tree
0.0177
0.0071
0.1792
0.7958
0.0002
0.0000
car
0.1755
0.0000
0.0211
0.0252
0.7782
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.772
0.903
0.819
0.808
0.433
--
Recall/Completeness
0.760
0.827
0.856
0.796
0.778
--
F1
0.766
0.863
0.837
0.802
0.557
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7938
0.0378
0.1205
0.0298
0.0181
0.0000
building
0.0788
0.8437
0.0632
0.0132
0.0010
0.0000
low_veg
0.0255
0.0164
0.8760
0.0812
0.0009
0.0000
tree
0.0118
0.0041
0.1608
0.8232
0.0001
0.0000
car
0.1269
0.0000
0.0145
0.0189
0.8397
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.797
0.924
0.840
0.834
0.416
--
Recall/Completeness
0.794
0.844
0.876
0.823
0.840
--
F1
0.795
0.882
0.858
0.829
0.557
--

Overall accuracy 0.845

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
0.8764
0.0387
0.0274
0.0095
0.0479
0.0000
building
0.1175
0.8446
0.0259
0.0067
0.0051
0.0003
low_veg
0.0787
0.0388
0.6734
0.2043
0.0042
0.0006
tree
0.0140
0.0075
0.0709
0.9059
0.0017
0.0000
car
0.1989
0.0224
0.0106
0.0076
0.7605
0.0000
clutter
0.4317
0.2321
0.1469
0.0087
0.1805
0.0000
Precision/Correctness
0.811
0.931
0.837
0.666
0.572
0.000
Recall/Completeness
0.876
0.845
0.673
0.906
0.760
0.000
F1
0.842
0.885
0.746
0.767
0.653
-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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9149
0.0293
0.0196
0.0043
0.0318
0.0000
building
0.1070
0.8648
0.0197
0.0037
0.0045
0.0003
low_veg
0.0656
0.0261
0.7110
0.1938
0.0028
0.0006
tree
0.0067
0.0035
0.0533
0.9357
0.0008
0.0000
car
0.1542
0.0217
0.0086
0.0054
0.8101
0.0000
clutter
0.4277
0.2495
0.1308
0.0058
0.1862
0.0000
Precision/Correctness
0.837
0.953
0.876
0.698
0.610
0.000
Recall/Completeness
0.915
0.865
0.711
0.936
0.810
0.000
F1
0.874
0.907
0.785
0.800
0.696
-nan

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8874
0.0420
0.0222
0.0157
0.0321
0.0006
building
0.0734
0.8999
0.0116
0.0086
0.0015
0.0050
low_veg
0.0661
0.0428
0.6207
0.2634
0.0070
0.0001
tree
0.0087
0.0063
0.0448
0.9397
0.0005
0.0000
car
0.1328
0.0393
0.0018
0.0103
0.8159
0.0000
clutter
0.2669
0.5151
0.0336
0.0138
0.1502
0.0204
Precision/Correctness
0.903
0.858
0.833
0.679
0.358
0.274
Recall/Completeness
0.887
0.900
0.621
0.940
0.816
0.020
F1
0.895
0.878
0.711
0.789
0.498
0.038

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9147
0.0340
0.0163
0.0081
0.0264
0.0005
building
0.0692
0.9122
0.0074
0.0048
0.0011
0.0052
low_veg
0.0470
0.0287
0.6673
0.2517
0.0053
0.0000
tree
0.0037
0.0039
0.0345
0.9578
0.0001
0.0000
car
0.0964
0.0383
0.0006
0.0069
0.8578
0.0000
clutter
0.2424
0.5515
0.0280
0.0063
0.1514
0.0203
Precision/Correctness
0.921
0.882
0.871
0.720
0.347
0.249
Recall/Completeness
0.915
0.912
0.667
0.958
0.858
0.020
F1
0.918
0.897
0.756
0.822
0.494
0.038

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8506
0.0635
0.0155
0.0063
0.0634
0.0008
building
0.4240
0.5486
0.0111
0.0033
0.0127
0.0003
low_veg
0.0536
0.0123
0.9015
0.0247
0.0076
0.0003
tree
0.0374
0.0067
0.1188
0.8304
0.0065
0.0002
car
0.1380
0.2098
0.0078
0.0053
0.6319
0.0072
clutter
0.2819
0.4615
0.0231
0.0160
0.2105
0.0069
Precision/Correctness
0.697
0.751
0.850
0.924
0.380
0.250
Recall/Completeness
0.851
0.549
0.902
0.830
0.632
0.007
F1
0.766
0.634
0.875
0.875
0.475
0.014

Overall accuracy 0.736

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
0.8780
0.0574
0.0121
0.0028
0.0492
0.0005
building
0.4278
0.5538
0.0065
0.0010
0.0108
0.0002
low_veg
0.0346
0.0085
0.9323
0.0191
0.0054
0.0001
tree
0.0163
0.0039
0.1004
0.8764
0.0030
0.0000
car
0.1140
0.2357
0.0043
0.0024
0.6355
0.0082
clutter
0.2496
0.5131
0.0205
0.0078
0.2022
0.0068
Precision/Correctness
0.707
0.776
0.884
0.955
0.382
0.286
Recall/Completeness
0.878
0.554
0.932
0.876
0.636
0.007
F1
0.783
0.646
0.907
0.914
0.477
0.013

Overall accuracy 0.758

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
0.9059
0.0520
0.0125
0.0046
0.0202
0.0048
building
0.1358
0.7190
0.1172
0.0184
0.0081
0.0014
low_veg
0.0743
0.0250
0.7340
0.1601
0.0047
0.0020
tree
0.0413
0.0093
0.1912
0.7565
0.0014
0.0003
car
0.1985
0.0803
0.0173
0.0005
0.7033
0.0000
clutter
0.3361
0.4273
0.0070
0.0001
0.2243
0.0052
Precision/Correctness
0.821
0.810
0.655
0.714
0.281
0.097
Recall/Completeness
0.906
0.719
0.734
0.757
0.703
0.005
F1
0.862
0.762
0.693
0.735
0.401
0.010

Overall accuracy 0.758

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
0.9268
0.0450
0.0072
0.0019
0.0143
0.0048
building
0.1332
0.7262
0.1148
0.0173
0.0071
0.0014
low_veg
0.0478
0.0173
0.7616
0.1685
0.0031
0.0017
tree
0.0187
0.0046
0.1787
0.7973
0.0006
0.0001
car
0.1510
0.0776
0.0135
0.0001
0.7577
0.0000
clutter
0.3050
0.4498
0.0056
0.0000
0.2345
0.0050
Precision/Correctness
0.844
0.834
0.673
0.722
0.274
0.089
Recall/Completeness
0.927
0.726
0.762
0.797
0.758
0.005
F1
0.883
0.776
0.714
0.758
0.403
0.009

Overall accuracy 0.780

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
0.8227
0.0443
0.0889
0.0255
0.0183
0.0003
building
0.0926
0.8633
0.0275
0.0089
0.0075
0.0001
low_veg
0.0396
0.0217
0.7129
0.2219
0.0038
0.0001
tree
0.0289
0.0064
0.1607
0.8017
0.0019
0.0003
car
0.2504
0.0319
0.0255
0.0107
0.6810
0.0004
clutter
--
--
--
--
--
--
Precision/Correctness
0.827
0.926
0.724
0.746
0.391
--
Recall/Completeness
0.823
0.863
0.713
0.802
0.681
--
F1
0.825
0.894
0.718
0.773
0.497
--

Overall accuracy 0.800

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
0.8743
0.0364
0.0635
0.0105
0.0151
0.0002
building
0.0850
0.8891
0.0156
0.0034
0.0068
0.0000
low_veg
0.0283
0.0136
0.7389
0.2158
0.0033
0.0000
tree
0.0112
0.0028
0.1329
0.8518
0.0010
0.0003
car
0.1974
0.0298
0.0246
0.0041
0.7440
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.862
0.950
0.782
0.775
0.395
--
Recall/Completeness
0.874
0.889
0.739
0.852
0.744
--
F1
0.868
0.919
0.760
0.812
0.516
--

Overall accuracy 0.840

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
0.8742
0.0403
0.0476
0.0167
0.0206
0.0005
building
0.1016
0.8269
0.0566
0.0090
0.0053
0.0007
low_veg
0.0568
0.0331
0.8175
0.0887
0.0038
0.0001
tree
0.0261
0.0088
0.1878
0.7763
0.0009
0.0002
car
0.1681
0.0884
0.0296
0.0055
0.7083
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.829
0.902
0.781
0.827
0.457
--
Recall/Completeness
0.874
0.827
0.817
0.776
0.708
--
F1
0.851
0.863
0.799
0.801
0.555
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9064
0.0368
0.0337
0.0069
0.0158
0.0004
building
0.0962
0.8489
0.0452
0.0046
0.0045
0.0006
low_veg
0.0415
0.0263
0.8529
0.0765
0.0028
0.0001
tree
0.0123
0.0059
0.1654
0.8159
0.0004
0.0002
car
0.1234
0.0986
0.0243
0.0013
0.7524
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.859
0.922
0.818
0.865
0.467
--
Recall/Completeness
0.906
0.849
0.853
0.816
0.752
--
F1
0.882
0.884
0.835
0.840
0.576
--

Overall accuracy 0.858

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image