Detailled semantic labeling (2D) result


Name K. Takahashi
Affiliation Waseda University, Japan
Abbreviation WAS1
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
0.890
0.037
0.050
0.017
0.006
0.000
building
0.081
0.881
0.030
0.005
0.001
0.002
low_veg
0.045
0.030
0.743
0.181
0.000
0.000
tree
0.021
0.006
0.083
0.890
0.000
0.000
car
0.313
0.019
0.008
0.005
0.656
0.000
clutter
0.525
0.249
0.010
0.003
0.086
0.127
Precision/Correctness
0.848
0.918
0.799
0.816
0.748
0.707
Recall/Completeness
0.890
0.881
0.743
0.890
0.656
0.127
F1
0.868
0.899
0.770
0.851
0.699
0.215

Overall accuracy 0.846

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
0.915
0.031
0.041
0.009
0.004
0.000
building
0.073
0.900
0.022
0.002
0.001
0.002
low_veg
0.034
0.023
0.778
0.165
0.000
0.000
tree
0.010
0.002
0.068
0.920
0.000
0.000
car
0.205
0.020
0.006
0.002
0.766
0.000
clutter
0.499
0.263
0.007
0.001
0.090
0.140
Precision/Correctness
0.878
0.936
0.834
0.846
0.774
0.713
Recall/Completeness
0.915
0.900
0.778
0.920
0.766
0.140
F1
0.896
0.918
0.805
0.881
0.770
0.234

Overall accuracy 0.875
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9334
0.0142
0.0295
0.0147
0.0082
0.0000
building
0.0760
0.8849
0.0339
0.0041
0.0010
0.0000
low_veg
0.0486
0.0232
0.6951
0.2330
0.0001
0.0000
tree
0.0278
0.0042
0.0661
0.8991
0.0003
0.0025
car
0.3283
0.0005
0.0000
0.0008
0.6703
0.0000
clutter
0.1844
0.0000
0.0000
0.0059
0.0000
0.8097
Precision/Correctness
0.839
0.952
0.844
0.782
0.815
0.960
Recall/Completeness
0.933
0.885
0.695
0.899
0.670
0.810
F1
0.884
0.917
0.762
0.837
0.736
0.878

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9608
0.0106
0.0176
0.0050
0.0059
0.0000
building
0.0698
0.9063
0.0218
0.0010
0.0010
0.0000
low_veg
0.0332
0.0159
0.7248
0.2260
0.0000
0.0000
tree
0.0128
0.0015
0.0540
0.9302
0.0002
0.0014
car
0.2060
0.0000
0.0000
0.0002
0.7939
0.0000
clutter
0.1910
0.0000
0.0000
0.0008
0.0000
0.8081
Precision/Correctness
0.875
0.969
0.883
0.807
0.844
0.979
Recall/Completeness
0.961
0.906
0.725
0.930
0.794
0.808
F1
0.916
0.937
0.796
0.864
0.818
0.885

Overall accuracy 0.879

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9313
0.0412
0.0159
0.0049
0.0067
0.0000
building
0.0840
0.8886
0.0122
0.0021
0.0003
0.0128
low_veg
0.1211
0.0387
0.4959
0.3397
0.0046
0.0000
tree
0.0422
0.0140
0.0719
0.8713
0.0006
0.0000
car
0.4022
0.0242
0.0001
0.0009
0.5726
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.818
0.971
0.596
0.702
0.778
--
Recall/Completeness
0.931
0.889
0.496
0.871
0.573
--
F1
0.871
0.928
0.541
0.778
0.660
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9542
0.0259
0.0123
0.0025
0.0050
0.0000
building
0.0678
0.9071
0.0106
0.0008
0.0003
0.0134
low_veg
0.0784
0.0117
0.5313
0.3741
0.0045
0.0000
tree
0.0147
0.0059
0.0638
0.9155
0.0001
0.0000
car
0.2936
0.0224
0.0000
0.0000
0.6839
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.853
0.985
0.626
0.727
0.807
--
Recall/Completeness
0.954
0.907
0.531
0.915
0.684
--
F1
0.901
0.944
0.575
0.810
0.740
--

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_area6_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9065
0.0373
0.0370
0.0114
0.0078
0.0000
building
0.0709
0.9047
0.0207
0.0034
0.0002
0.0000
low_veg
0.1019
0.0182
0.6675
0.2122
0.0002
0.0000
tree
0.0177
0.0045
0.0602
0.9174
0.0002
0.0000
car
0.4399
0.0099
0.0015
0.0012
0.5475
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.832
0.959
0.679
0.908
0.808
--
Recall/Completeness
0.906
0.905
0.668
0.917
0.548
--
F1
0.868
0.931
0.673
0.913
0.653
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9355
0.0261
0.0286
0.0046
0.0053
0.0000
building
0.0571
0.9260
0.0161
0.0007
0.0002
0.0000
low_veg
0.0825
0.0066
0.7126
0.1983
0.0000
0.0000
tree
0.0063
0.0014
0.0496
0.9427
0.0001
0.0000
car
0.3395
0.0072
0.0010
0.0000
0.6522
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.874
0.977
0.717
0.931
0.852
--
Recall/Completeness
0.935
0.926
0.713
0.943
0.652
--
F1
0.904
0.951
0.715
0.937
0.739
--

Overall accuracy 0.909

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8644
0.0314
0.0879
0.0106
0.0056
0.0001
building
0.0538
0.9084
0.0327
0.0042
0.0008
0.0000
low_veg
0.0395
0.0085
0.8372
0.1145
0.0003
0.0000
tree
0.0229
0.0062
0.0730
0.8972
0.0001
0.0006
car
0.3486
0.0093
0.0249
0.0027
0.6145
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.896
0.941
0.756
0.889
0.837
--
Recall/Completeness
0.864
0.908
0.837
0.897
0.614
--
F1
0.880
0.924
0.795
0.893
0.709
--

Overall accuracy 0.873

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8900
0.0264
0.0760
0.0041
0.0035
0.0000
building
0.0440
0.9315
0.0224
0.0013
0.0008
0.0000
low_veg
0.0239
0.0026
0.8701
0.1034
0.0000
0.0000
tree
0.0084
0.0028
0.0545
0.9337
0.0000
0.0005
car
0.2521
0.0074
0.0208
0.0021
0.7176
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.932
0.959
0.799
0.912
0.876
--
Recall/Completeness
0.890
0.932
0.870
0.934
0.718
--
F1
0.910
0.945
0.833
0.923
0.789
--

Overall accuracy 0.904

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
0.8270
0.0748
0.0717
0.0212
0.0053
0.0000
building
0.0795
0.8893
0.0170
0.0135
0.0006
0.0000
low_veg
0.0442
0.0508
0.6756
0.2294
0.0000
0.0000
tree
0.0158
0.0078
0.0682
0.9082
0.0000
0.0000
car
0.2318
0.0268
0.0142
0.0020
0.7249
0.0002
clutter
--
--
--
--
--
--
Precision/Correctness
0.908
0.844
0.725
0.784
0.708
--
Recall/Completeness
0.827
0.889
0.676
0.908
0.725
--
F1
0.866
0.866
0.699
0.841
0.716
--

Overall accuracy 0.831

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
0.8562
0.0659
0.0635
0.0102
0.0041
0.0000
building
0.0742
0.9028
0.0126
0.0098
0.0005
0.0000
low_veg
0.0398
0.0369
0.7218
0.2014
0.0000
0.0000
tree
0.0070
0.0038
0.0523
0.9369
0.0000
0.0000
car
0.1284
0.0306
0.0135
0.0000
0.8275
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.925
0.870
0.762
0.823
0.720
--
Recall/Completeness
0.856
0.903
0.722
0.937
0.828
--
F1
0.889
0.886
0.742
0.876
0.770
--

Overall accuracy 0.861

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
0.8806
0.0338
0.0484
0.0328
0.0043
0.0001
building
0.1381
0.8452
0.0108
0.0059
0.0000
0.0000
low_veg
0.0483
0.0326
0.7013
0.2176
0.0002
0.0000
tree
0.0114
0.0074
0.0818
0.8993
0.0000
0.0000
car
0.2208
0.0018
0.0000
0.0027
0.7747
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.732
0.905
0.838
0.828
0.854
--
Recall/Completeness
0.881
0.845
0.701
0.899
0.775
--
F1
0.800
0.874
0.764
0.862
0.812
--

Overall accuracy 0.828

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
0.9084
0.0299
0.0407
0.0187
0.0022
0.0001
building
0.1420
0.8512
0.0054
0.0013
0.0000
0.0000
low_veg
0.0457
0.0214
0.7416
0.1912
0.0001
0.0000
tree
0.0055
0.0029
0.0688
0.9228
0.0000
0.0000
car
0.0854
0.0005
0.0000
0.0000
0.9141
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.751
0.935
0.866
0.857
0.909
--
Recall/Completeness
0.908
0.851
0.742
0.923
0.914
--
F1
0.822
0.891
0.799
0.889
0.912
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8215
0.0316
0.1042
0.0368
0.0058
0.0001
building
0.0250
0.9504
0.0182
0.0065
0.0000
0.0000
low_veg
0.0774
0.0462
0.6562
0.2190
0.0012
0.0000
tree
0.0231
0.0129
0.0916
0.8724
0.0000
0.0000
car
0.2045
0.0000
0.0000
0.0081
0.7874
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.853
0.878
0.743
0.817
0.561
--
Recall/Completeness
0.822
0.950
0.656
0.872
0.787
--
F1
0.837
0.913
0.697
0.844
0.655
--

Overall accuracy 0.819

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
0.8403
0.0272
0.1015
0.0259
0.0051
0.0000
building
0.0191
0.9695
0.0096
0.0018
0.0000
0.0000
low_veg
0.0719
0.0324
0.6852
0.2092
0.0013
0.0000
tree
0.0151
0.0065
0.0805
0.8979
0.0000
0.0000
car
0.0940
0.0000
0.0000
0.0040
0.9020
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.881
0.910
0.764
0.839
0.538
--
Recall/Completeness
0.840
0.970
0.685
0.898
0.902
--
F1
0.860
0.939
0.723
0.867
0.674
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8154
0.0388
0.0871
0.0560
0.0028
0.0000
building
0.0734
0.8941
0.0216
0.0108
0.0002
0.0000
low_veg
0.0174
0.0172
0.8254
0.1400
0.0000
0.0000
tree
0.0257
0.0059
0.0969
0.8715
0.0000
0.0000
car
0.2776
0.0043
0.0000
0.0013
0.7169
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.817
0.923
0.822
0.860
0.774
--
Recall/Completeness
0.815
0.894
0.825
0.872
0.717
--
F1
0.816
0.908
0.824
0.866
0.745
--

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_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8476
0.0339
0.0811
0.0361
0.0013
0.0000
building
0.0710
0.9136
0.0124
0.0028
0.0002
0.0000
low_veg
0.0130
0.0112
0.8517
0.1242
0.0000
0.0000
tree
0.0171
0.0026
0.0841
0.8962
0.0000
0.0000
car
0.1835
0.0001
0.0000
0.0000
0.8163
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.847
0.946
0.848
0.885
0.844
--
Recall/Completeness
0.848
0.914
0.852
0.896
0.816
--
F1
0.847
0.929
0.850
0.891
0.830
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8438
0.0594
0.0460
0.0463
0.0045
0.0000
building
0.0580
0.9086
0.0272
0.0062
0.0000
0.0000
low_veg
0.0307
0.0369
0.6721
0.2597
0.0005
0.0000
tree
0.0083
0.0016
0.0399
0.9502
0.0000
0.0000
car
0.1479
0.0688
0.0306
0.0073
0.7454
0.0000
clutter
0.0000
0.9998
0.0000
0.0002
0.0000
0.0000
Precision/Correctness
0.897
0.759
0.812
0.866
0.752
-nan
Recall/Completeness
0.844
0.909
0.672
0.950
0.745
0.000
F1
0.870
0.827
0.735
0.906
0.749
-nan

Overall accuracy 0.842

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8745
0.0507
0.0434
0.0283
0.0030
0.0000
building
0.0560
0.9199
0.0227
0.0014
0.0000
0.0000
low_veg
0.0243
0.0250
0.7203
0.2299
0.0005
0.0000
tree
0.0033
0.0003
0.0305
0.9659
0.0000
0.0000
car
0.0483
0.0643
0.0253
0.0017
0.8605
0.0000
clutter
0.0000
1.0000
0.0000
0.0000
0.0000
0.0000
Precision/Correctness
0.920
0.779
0.840
0.893
0.783
-nan
Recall/Completeness
0.875
0.920
0.720
0.966
0.860
0.000
F1
0.896
0.844
0.776
0.928
0.820
-nan

Overall accuracy 0.867

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
0.8870
0.0245
0.0458
0.0355
0.0073
0.0000
building
0.0653
0.9118
0.0148
0.0077
0.0004
0.0000
low_veg
0.0686
0.0450
0.6649
0.2213
0.0001
0.0000
tree
0.0230
0.0040
0.0474
0.9252
0.0005
0.0000
car
0.2723
0.0130
0.0049
0.0127
0.6971
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.890
0.899
0.812
0.823
0.750
--
Recall/Completeness
0.887
0.912
0.665
0.925
0.697
--
F1
0.888
0.905
0.731
0.871
0.723
--

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_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9191
0.0172
0.0394
0.0193
0.0050
0.0000
building
0.0605
0.9280
0.0082
0.0031
0.0003
0.0000
low_veg
0.0662
0.0314
0.7128
0.1896
0.0000
0.0000
tree
0.0103
0.0011
0.0356
0.9527
0.0003
0.0000
car
0.1722
0.0171
0.0025
0.0041
0.8041
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.913
0.933
0.850
0.862
0.788
--
Recall/Completeness
0.919
0.928
0.713
0.953
0.804
--
F1
0.916
0.930
0.776
0.905
0.796
--

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_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7722
0.0540
0.1332
0.0312
0.0093
0.0000
building
0.0606
0.8484
0.0852
0.0058
0.0000
0.0000
low_veg
0.0201
0.0226
0.8670
0.0902
0.0001
0.0000
tree
0.0152
0.0078
0.1507
0.8262
0.0000
0.0000
car
0.2586
0.0000
0.0005
0.0028
0.7381
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.821
0.893
0.829
0.831
0.663
--
Recall/Completeness
0.772
0.848
0.867
0.826
0.738
--
F1
0.796
0.870
0.847
0.828
0.698
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8046
0.0441
0.1235
0.0200
0.0078
0.0000
building
0.0595
0.8576
0.0809
0.0021
0.0000
0.0000
low_veg
0.0177
0.0173
0.8873
0.0776
0.0002
0.0000
tree
0.0093
0.0033
0.1314
0.8560
0.0000
0.0000
car
0.1630
0.0000
0.0000
0.0000
0.8370
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.843
0.919
0.849
0.857
0.658
--
Recall/Completeness
0.805
0.858
0.887
0.856
0.837
--
F1
0.823
0.887
0.868
0.856
0.737
--

Overall accuracy 0.862

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9245
0.0300
0.0284
0.0070
0.0102
0.0000
building
0.1181
0.8501
0.0272
0.0044
0.0002
0.0000
low_veg
0.0625
0.0198
0.6328
0.2848
0.0001
0.0000
tree
0.0115
0.0037
0.0451
0.9396
0.0002
0.0000
car
0.3308
0.0034
0.0070
0.0045
0.6543
0.0000
clutter
0.5559
0.1955
0.1071
0.0068
0.1347
0.0000
Precision/Correctness
0.816
0.954
0.840
0.612
0.851
-nan
Recall/Completeness
0.924
0.850
0.633
0.940
0.654
0.000
F1
0.867
0.899
0.722
0.741
0.740
-nan

Overall accuracy 0.832

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
0.9495
0.0218
0.0209
0.0026
0.0052
0.0000
building
0.1067
0.8714
0.0202
0.0015
0.0002
0.0000
low_veg
0.0434
0.0097
0.6692
0.2776
0.0001
0.0000
tree
0.0036
0.0010
0.0346
0.9608
0.0000
0.0000
car
0.2232
0.0014
0.0046
0.0032
0.7676
0.0000
clutter
0.5619
0.1867
0.0908
0.0035
0.1571
0.0000
Precision/Correctness
0.848
0.972
0.877
0.635
0.903
-nan
Recall/Completeness
0.949
0.871
0.669
0.961
0.768
0.000
F1
0.896
0.919
0.759
0.765
0.830
-nan

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9352
0.0210
0.0256
0.0128
0.0054
0.0000
building
0.0507
0.9305
0.0120
0.0067
0.0001
0.0000
low_veg
0.0608
0.0199
0.5667
0.3526
0.0000
0.0000
tree
0.0062
0.0016
0.0209
0.9713
0.0000
0.0000
car
0.3075
0.0028
0.0015
0.0159
0.6724
0.0000
clutter
0.6294
0.2119
0.0245
0.0037
0.1305
0.0000
Precision/Correctness
0.896
0.932
0.830
0.640
0.593
-nan
Recall/Completeness
0.935
0.931
0.567
0.971
0.672
0.000
F1
0.915
0.931
0.674
0.772
0.630
-nan

Overall accuracy 0.852

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9553
0.0160
0.0191
0.0054
0.0042
0.0000
building
0.0434
0.9466
0.0067
0.0032
0.0001
0.0000
low_veg
0.0394
0.0080
0.6066
0.3461
0.0000
0.0000
tree
0.0019
0.0001
0.0142
0.9838
0.0000
0.0000
car
0.2003
0.0009
0.0014
0.0138
0.7836
0.0000
clutter
0.6207
0.2174
0.0161
0.0008
0.1450
0.0000
Precision/Correctness
0.919
0.951
0.873
0.673
0.582
-nan
Recall/Completeness
0.955
0.947
0.607
0.984
0.784
0.000
F1
0.937
0.949
0.716
0.800
0.668
-nan

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9234
0.0541
0.0092
0.0031
0.0102
0.0000
building
0.1737
0.8051
0.0120
0.0027
0.0065
0.0000
low_veg
0.0554
0.0014
0.9042
0.0388
0.0002
0.0000
tree
0.0585
0.0028
0.0636
0.8747
0.0004
0.0000
car
0.2291
0.0653
0.0021
0.0042
0.6993
0.0000
clutter
0.6349
0.2271
0.0084
0.0058
0.1239
0.0000
Precision/Correctness
0.795
0.876
0.903
0.925
0.716
-nan
Recall/Completeness
0.923
0.805
0.904
0.875
0.699
0.000
F1
0.854
0.839
0.904
0.899
0.707
-nan

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9333
0.0539
0.0068
0.0008
0.0052
0.0000
building
0.1625
0.8248
0.0066
0.0004
0.0057
0.0000
low_veg
0.0351
0.0002
0.9325
0.0322
0.0000
0.0000
tree
0.0237
0.0005
0.0471
0.9286
0.0001
0.0000
car
0.1335
0.0742
0.0008
0.0012
0.7902
0.0000
clutter
0.6072
0.2497
0.0086
0.0012
0.1333
0.0000
Precision/Correctness
0.825
0.886
0.933
0.951
0.746
-nan
Recall/Completeness
0.933
0.825
0.932
0.929
0.790
0.000
F1
0.876
0.854
0.933
0.939
0.768
-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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9579
0.0247
0.0111
0.0035
0.0028
0.0000
building
0.1013
0.8262
0.0655
0.0025
0.0046
0.0000
low_veg
0.0576
0.0120
0.8789
0.0514
0.0001
0.0000
tree
0.0444
0.0056
0.1322
0.8179
0.0000
0.0000
car
0.3272
0.0196
0.0018
0.0018
0.6496
0.0000
clutter
0.8425
0.0334
0.0058
0.0006
0.1177
0.0000
Precision/Correctness
0.799
0.945
0.783
0.895
0.487
-nan
Recall/Completeness
0.958
0.826
0.879
0.818
0.650
0.000
F1
0.871
0.881
0.828
0.855
0.557
-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_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9693
0.0229
0.0053
0.0012
0.0012
0.0000
building
0.0915
0.8433
0.0604
0.0006
0.0042
0.0000
low_veg
0.0316
0.0081
0.9140
0.0463
0.0001
0.0000
tree
0.0182
0.0036
0.1199
0.8583
0.0000
0.0000
car
0.1982
0.0184
0.0013
0.0005
0.7817
0.0000
clutter
0.8327
0.0322
0.0053
0.0001
0.1297
0.0000
Precision/Correctness
0.828
0.954
0.809
0.920
0.485
-nan
Recall/Completeness
0.969
0.843
0.914
0.858
0.782
0.000
F1
0.893
0.895
0.858
0.888
0.598
-nan

Overall accuracy 0.861

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
0.8046
0.0527
0.1171
0.0222
0.0033
0.0001
building
0.0407
0.9181
0.0333
0.0075
0.0004
0.0000
low_veg
0.0219
0.0112
0.6945
0.2723
0.0000
0.0002
tree
0.0249
0.0043
0.1073
0.8633
0.0002
0.0000
car
0.3198
0.0048
0.0010
0.0077
0.6667
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.892
0.935
0.731
0.729
0.839
--
Recall/Completeness
0.805
0.918
0.694
0.863
0.667
--
F1
0.846
0.926
0.712
0.790
0.743
--

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_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8544
0.0470
0.0886
0.0077
0.0023
0.0001
building
0.0328
0.9459
0.0194
0.0016
0.0004
0.0000
low_veg
0.0133
0.0045
0.7169
0.2652
0.0000
0.0001
tree
0.0069
0.0010
0.0804
0.9116
0.0000
0.0000
car
0.1960
0.0033
0.0000
0.0017
0.7990
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.932
0.954
0.794
0.756
0.874
--
Recall/Completeness
0.854
0.946
0.717
0.912
0.799
--
F1
0.892
0.950
0.753
0.826
0.835
--

Overall accuracy 0.859

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
0.9051
0.0280
0.0517
0.0113
0.0039
0.0000
building
0.0675
0.8661
0.0624
0.0032
0.0009
0.0000
low_veg
0.0323
0.0640
0.8206
0.0828
0.0002
0.0000
tree
0.0228
0.0032
0.1459
0.8281
0.0001
0.0000
car
0.3402
0.0149
0.0346
0.0069
0.6034
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.878
0.893
0.797
0.856
0.811
--
Recall/Completeness
0.905
0.866
0.821
0.828
0.603
--
F1
0.891
0.879
0.809
0.842
0.692
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9322
0.0262
0.0357
0.0037
0.0022
0.0000
building
0.0604
0.8915
0.0465
0.0008
0.0008
0.0000
low_veg
0.0182
0.0651
0.8505
0.0660
0.0001
0.0000
tree
0.0071
0.0007
0.1225
0.8698
0.0000
0.0000
car
0.2338
0.0176
0.0324
0.0030
0.7132
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.912
0.903
0.838
0.894
0.859
--
Recall/Completeness
0.932
0.892
0.851
0.870
0.713
--
F1
0.922
0.897
0.844
0.882
0.779
--

Overall accuracy 0.886

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image