Detailled semantic labeling (2D) result


Name K. Takahashi
Affiliation Waseda University, Japan
Abbreviation WAS3
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
0.883
0.035
0.049
0.019
0.011
0.002
building
0.043
0.924
0.025
0.006
0.000
0.001
low_veg
0.047
0.027
0.773
0.152
0.001
0.000
tree
0.020
0.008
0.097
0.875
0.000
0.000
car
0.263
0.031
0.016
0.003
0.687
0.000
clutter
0.362
0.210
0.012
0.007
0.026
0.383
Precision/Correctness
0.882
0.924
0.798
0.833
0.713
0.782
Recall/Completeness
0.883
0.924
0.773
0.875
0.687
0.383
F1
0.883
0.924
0.785
0.853
0.700
0.515

Overall accuracy 0.861

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
0.914
0.027
0.040
0.010
0.007
0.001
building
0.035
0.943
0.018
0.002
0.000
0.001
low_veg
0.036
0.018
0.809
0.136
0.000
0.000
tree
0.010
0.003
0.080
0.906
0.000
0.000
car
0.179
0.033
0.012
0.001
0.774
0.000
clutter
0.341
0.213
0.009
0.004
0.026
0.407
Precision/Correctness
0.914
0.947
0.833
0.864
0.769
0.811
Recall/Completeness
0.914
0.943
0.809
0.906
0.774
0.407
F1
0.914
0.945
0.821
0.884
0.771
0.542

Overall accuracy 0.892
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9292
0.0145
0.0271
0.0155
0.0136
0.0000
building
0.0431
0.9214
0.0276
0.0063
0.0002
0.0013
low_veg
0.0604
0.0189
0.7003
0.2202
0.0001
0.0000
tree
0.0270
0.0052
0.0677
0.8957
0.0002
0.0041
car
0.2419
0.0008
0.0037
0.0005
0.7531
0.0000
clutter
0.0000
0.0000
0.0000
0.0033
0.0000
0.9967
Precision/Correctness
0.869
0.957
0.851
0.789
0.768
0.933
Recall/Completeness
0.929
0.921
0.700
0.896
0.753
0.997
F1
0.898
0.939
0.769
0.839
0.760
0.964

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9596
0.0104
0.0160
0.0053
0.0088
0.0000
building
0.0358
0.9445
0.0164
0.0016
0.0003
0.0014
low_veg
0.0443
0.0099
0.7312
0.2145
0.0000
0.0000
tree
0.0123
0.0019
0.0552
0.9277
0.0001
0.0028
car
0.1355
0.0000
0.0026
0.0000
0.8619
0.0000
clutter
0.0000
0.0000
0.0000
0.0008
0.0000
0.9992
Precision/Correctness
0.909
0.976
0.890
0.813
0.819
0.950
Recall/Completeness
0.960
0.944
0.731
0.928
0.862
0.999
F1
0.933
0.960
0.803
0.867
0.840
0.974

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9223
0.0477
0.0158
0.0047
0.0095
0.0000
building
0.0746
0.9121
0.0111
0.0016
0.0007
0.0000
low_veg
0.1586
0.0573
0.5043
0.2751
0.0047
0.0000
tree
0.0433
0.0233
0.0580
0.8746
0.0007
0.0000
car
0.2979
0.0114
0.0015
0.0005
0.6887
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.829
0.966
0.622
0.742
0.746
--
Recall/Completeness
0.922
0.912
0.504
0.875
0.689
--
F1
0.873
0.938
0.557
0.803
0.716
--

Overall accuracy 0.891

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
0.9521
0.0295
0.0121
0.0015
0.0048
0.0000
building
0.0587
0.9306
0.0096
0.0004
0.0006
0.0000
low_veg
0.1194
0.0357
0.5388
0.3035
0.0027
0.0000
tree
0.0175
0.0112
0.0486
0.9227
0.0000
0.0000
car
0.2095
0.0095
0.0003
0.0000
0.7807
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.865
0.981
0.654
0.771
0.822
--
Recall/Completeness
0.952
0.931
0.539
0.923
0.781
--
F1
0.906
0.955
0.591
0.840
0.801
--

Overall accuracy 0.919

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
0.8698
0.0660
0.0468
0.0071
0.0102
0.0001
building
0.0414
0.9384
0.0177
0.0024
0.0002
0.0000
low_veg
0.0878
0.0282
0.7266
0.1568
0.0007
0.0000
tree
0.0178
0.0066
0.0845
0.8911
0.0001
0.0000
car
0.3191
0.0198
0.0083
0.0006
0.6522
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.869
0.934
0.647
0.930
0.794
--
Recall/Completeness
0.870
0.938
0.727
0.891
0.652
--
F1
0.869
0.936
0.685
0.910
0.716
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9062
0.0509
0.0363
0.0014
0.0051
0.0001
building
0.0305
0.9561
0.0128
0.0005
0.0001
0.0000
low_veg
0.0719
0.0148
0.7690
0.1440
0.0002
0.0000
tree
0.0086
0.0022
0.0678
0.9214
0.0000
0.0000
car
0.2310
0.0155
0.0042
0.0001
0.7492
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.908
0.956
0.690
0.950
0.874
--
Recall/Completeness
0.906
0.956
0.769
0.921
0.749
--
F1
0.907
0.956
0.727
0.935
0.807
--

Overall accuracy 0.913

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
0.8672
0.0290
0.0812
0.0108
0.0105
0.0014
building
0.0349
0.9361
0.0257
0.0034
0.0000
0.0000
low_veg
0.0320
0.0161
0.8623
0.0893
0.0003
0.0000
tree
0.0234
0.0078
0.0766
0.8915
0.0000
0.0007
car
0.2578
0.0115
0.0488
0.0017
0.6802
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.918
0.938
0.770
0.907
0.775
--
Recall/Completeness
0.867
0.936
0.862
0.891
0.680
--
F1
0.892
0.937
0.814
0.899
0.724
--

Overall accuracy 0.885

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
0.9004
0.0212
0.0681
0.0034
0.0055
0.0013
building
0.0249
0.9580
0.0163
0.0009
0.0000
0.0000
low_veg
0.0177
0.0089
0.8949
0.0785
0.0001
0.0000
tree
0.0090
0.0035
0.0555
0.9317
0.0000
0.0004
car
0.1795
0.0107
0.0370
0.0008
0.7721
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.953
0.961
0.817
0.931
0.850
--
Recall/Completeness
0.900
0.958
0.895
0.932
0.772
--
F1
0.926
0.959
0.854
0.931
0.809
--

Overall accuracy 0.919

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
0.8166
0.0823
0.0733
0.0215
0.0062
0.0000
building
0.0368
0.9345
0.0173
0.0099
0.0015
0.0000
low_veg
0.0597
0.0540
0.7247
0.1611
0.0006
0.0000
tree
0.0159
0.0092
0.1091
0.8658
0.0000
0.0000
car
0.2484
0.0522
0.0065
0.0017
0.6913
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.927
0.838
0.702
0.821
0.636
--
Recall/Completeness
0.817
0.935
0.725
0.866
0.691
--
F1
0.868
0.884
0.713
0.843
0.663
--

Overall accuracy 0.838

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
0.8463
0.0714
0.0671
0.0107
0.0045
0.0000
building
0.0278
0.9522
0.0116
0.0070
0.0015
0.0000
low_veg
0.0563
0.0422
0.7657
0.1352
0.0006
0.0000
tree
0.0061
0.0042
0.0890
0.9006
0.0000
0.0000
car
0.1646
0.0526
0.0021
0.0000
0.7806
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.947
0.866
0.736
0.863
0.654
--
Recall/Completeness
0.846
0.952
0.766
0.901
0.781
--
F1
0.894
0.907
0.751
0.881
0.712
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8694
0.0465
0.0456
0.0321
0.0065
0.0000
building
0.0343
0.9457
0.0151
0.0048
0.0000
0.0000
low_veg
0.0510
0.0296
0.7730
0.1452
0.0002
0.0011
tree
0.0121
0.0098
0.1255
0.8525
0.0000
0.0000
car
0.2686
0.0183
0.0030
0.0003
0.7098
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.820
0.905
0.801
0.869
0.784
--
Recall/Completeness
0.869
0.946
0.773
0.853
0.710
--
F1
0.844
0.925
0.787
0.861
0.745
--

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_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8981
0.0441
0.0361
0.0176
0.0041
0.0000
building
0.0314
0.9621
0.0054
0.0011
0.0000
0.0000
low_veg
0.0482
0.0193
0.8092
0.1219
0.0001
0.0012
tree
0.0066
0.0040
0.1100
0.8794
0.0000
0.0000
car
0.1374
0.0191
0.0005
0.0000
0.8430
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.850
0.934
0.830
0.898
0.834
--
Recall/Completeness
0.898
0.962
0.809
0.879
0.843
--
F1
0.873
0.948
0.819
0.888
0.838
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7841
0.0378
0.1337
0.0370
0.0074
0.0000
building
0.0126
0.9616
0.0204
0.0054
0.0000
0.0000
low_veg
0.0741
0.0568
0.7078
0.1610
0.0002
0.0000
tree
0.0245
0.0191
0.1313
0.8250
0.0000
0.0000
car
0.2157
0.0005
0.0013
0.0011
0.7814
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.858
0.851
0.697
0.846
0.540
--
Recall/Completeness
0.784
0.962
0.708
0.825
0.781
--
F1
0.819
0.903
0.702
0.835
0.639
--

Overall accuracy 0.810

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
0.8031
0.0326
0.1315
0.0262
0.0066
0.0000
building
0.0078
0.9799
0.0100
0.0023
0.0000
0.0000
low_veg
0.0705
0.0417
0.7367
0.1509
0.0002
0.0000
tree
0.0169
0.0113
0.1196
0.8522
0.0000
0.0000
car
0.1189
0.0000
0.0001
0.0000
0.8810
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.883
0.887
0.716
0.867
0.522
--
Recall/Completeness
0.803
0.980
0.737
0.852
0.881
--
F1
0.841
0.931
0.726
0.860
0.656
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8215
0.0334
0.0804
0.0599
0.0048
0.0000
building
0.0281
0.9383
0.0213
0.0122
0.0001
0.0000
low_veg
0.0217
0.0188
0.8289
0.1306
0.0000
0.0000
tree
0.0241
0.0077
0.1085
0.8597
0.0000
0.0000
car
0.2979
0.0011
0.0045
0.0039
0.6926
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.859
0.925
0.814
0.863
0.668
--
Recall/Completeness
0.822
0.938
0.829
0.860
0.693
--
F1
0.840
0.931
0.822
0.861
0.680
--

Overall accuracy 0.858

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8535
0.0283
0.0751
0.0400
0.0030
0.0000
building
0.0237
0.9604
0.0115
0.0044
0.0001
0.0000
low_veg
0.0175
0.0118
0.8546
0.1162
0.0000
0.0000
tree
0.0162
0.0040
0.0958
0.8841
0.0000
0.0000
car
0.2045
0.0000
0.0029
0.0006
0.7921
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.891
0.949
0.839
0.887
0.728
--
Recall/Completeness
0.854
0.960
0.855
0.884
0.792
--
F1
0.872
0.955
0.847
0.886
0.759
--

Overall accuracy 0.884

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
0.8634
0.0395
0.0328
0.0592
0.0052
0.0000
building
0.0170
0.9593
0.0167
0.0061
0.0000
0.0009
low_veg
0.0469
0.0343
0.7401
0.1787
0.0000
0.0000
tree
0.0095
0.0024
0.0454
0.9426
0.0000
0.0000
car
0.2104
0.0086
0.0245
0.0086
0.7478
0.0000
clutter
0.7148
0.2834
0.0010
0.0008
0.0000
0.0000
Precision/Correctness
0.829
0.871
0.836
0.891
0.746
0.000
Recall/Completeness
0.863
0.959
0.740
0.943
0.748
0.000
F1
0.846
0.913
0.785
0.916
0.747
-nan

Overall accuracy 0.864

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
0.8948
0.0323
0.0285
0.0415
0.0030
0.0000
building
0.0137
0.9714
0.0124
0.0015
0.0000
0.0010
low_veg
0.0404
0.0216
0.7882
0.1498
0.0000
0.0000
tree
0.0042
0.0005
0.0339
0.9613
0.0000
0.0000
car
0.1190
0.0071
0.0145
0.0029
0.8565
0.0000
clutter
0.7270
0.2728
0.0002
0.0000
0.0000
0.0000
Precision/Correctness
0.847
0.898
0.869
0.917
0.807
0.000
Recall/Completeness
0.895
0.971
0.788
0.961
0.856
0.000
F1
0.870
0.933
0.826
0.939
0.831
-nan

Overall accuracy 0.890

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
0.8882
0.0272
0.0426
0.0312
0.0108
0.0000
building
0.0407
0.9352
0.0170
0.0071
0.0001
0.0000
low_veg
0.0642
0.0443
0.7142
0.1767
0.0006
0.0000
tree
0.0220
0.0048
0.0585
0.9144
0.0003
0.0000
car
0.2604
0.0019
0.0239
0.0075
0.7064
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.906
0.898
0.810
0.849
0.679
--
Recall/Completeness
0.888
0.935
0.714
0.914
0.706
--
F1
0.897
0.916
0.759
0.880
0.692
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9213
0.0201
0.0351
0.0161
0.0074
0.0000
building
0.0325
0.9556
0.0090
0.0029
0.0000
0.0000
low_veg
0.0600
0.0312
0.7589
0.1495
0.0004
0.0000
tree
0.0095
0.0017
0.0409
0.9478
0.0001
0.0000
car
0.1651
0.0012
0.0150
0.0021
0.8166
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.932
0.930
0.856
0.886
0.725
--
Recall/Completeness
0.921
0.956
0.759
0.948
0.817
--
F1
0.927
0.943
0.805
0.916
0.768
--

Overall accuracy 0.902

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
0.7654
0.0400
0.1364
0.0462
0.0120
0.0000
building
0.0422
0.8595
0.0836
0.0146
0.0000
0.0000
low_veg
0.0201
0.0158
0.8734
0.0907
0.0001
0.0000
tree
0.0120
0.0079
0.1540
0.8259
0.0002
0.0000
car
0.2493
0.0000
0.0331
0.0033
0.7143
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.847
0.919
0.828
0.818
0.596
--
Recall/Completeness
0.765
0.859
0.873
0.826
0.714
--
F1
0.804
0.888
0.850
0.822
0.650
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8025
0.0292
0.1257
0.0336
0.0090
0.0000
building
0.0396
0.8710
0.0774
0.0120
0.0000
0.0000
low_veg
0.0181
0.0104
0.8928
0.0787
0.0000
0.0000
tree
0.0065
0.0037
0.1342
0.8554
0.0001
0.0000
car
0.1755
0.0000
0.0252
0.0009
0.7983
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.869
0.946
0.849
0.842
0.618
--
Recall/Completeness
0.803
0.871
0.893
0.855
0.798
--
F1
0.835
0.907
0.870
0.849
0.697
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9040
0.0343
0.0276
0.0091
0.0249
0.0000
building
0.0646
0.9070
0.0233
0.0038
0.0002
0.0010
low_veg
0.0682
0.0287
0.6835
0.2183
0.0013
0.0000
tree
0.0097
0.0057
0.0671
0.9175
0.0000
0.0000
car
0.2964
0.0113
0.0078
0.0048
0.6797
0.0000
clutter
0.3925
0.4146
0.1420
0.0039
0.0469
0.0000
Precision/Correctness
0.859
0.945
0.846
0.661
0.724
0.000
Recall/Completeness
0.904
0.907
0.684
0.918
0.680
0.000
F1
0.881
0.925
0.756
0.769
0.701
-nan

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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9384
0.0229
0.0205
0.0041
0.0141
0.0000
building
0.0533
0.9273
0.0170
0.0013
0.0001
0.0010
low_veg
0.0503
0.0158
0.7243
0.2086
0.0011
0.0000
tree
0.0031
0.0017
0.0490
0.9462
0.0000
0.0000
car
0.2123
0.0106
0.0043
0.0034
0.7695
0.0000
clutter
0.3734
0.4439
0.1300
0.0006
0.0521
0.0000
Precision/Correctness
0.895
0.967
0.885
0.691
0.797
0.000
Recall/Completeness
0.938
0.927
0.724
0.946
0.769
0.000
F1
0.916
0.947
0.796
0.799
0.783
-nan

Overall accuracy 0.888

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
0.9292
0.0171
0.0257
0.0153
0.0106
0.0020
building
0.0342
0.9354
0.0133
0.0062
0.0002
0.0106
low_veg
0.0556
0.0242
0.6352
0.2845
0.0004
0.0000
tree
0.0082
0.0040
0.0476
0.9401
0.0000
0.0000
car
0.2496
0.0338
0.0077
0.0063
0.7026
0.0000
clutter
0.3568
0.4386
0.0182
0.0146
0.0349
0.1369
Precision/Correctness
0.924
0.911
0.822
0.670
0.625
0.523
Recall/Completeness
0.929
0.935
0.635
0.940
0.703
0.137
F1
0.927
0.923
0.717
0.782
0.661
0.217

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9539
0.0103
0.0188
0.0076
0.0079
0.0016
building
0.0275
0.9501
0.0081
0.0031
0.0001
0.0111
low_veg
0.0348
0.0113
0.6788
0.2748
0.0002
0.0000
tree
0.0030
0.0019
0.0392
0.9560
0.0000
0.0000
car
0.1639
0.0399
0.0063
0.0033
0.7865
0.0000
clutter
0.3232
0.4676
0.0119
0.0072
0.0365
0.1536
Precision/Correctness
0.947
0.933
0.862
0.708
0.644
0.521
Recall/Completeness
0.954
0.950
0.679
0.956
0.787
0.154
F1
0.950
0.941
0.759
0.813
0.708
0.237

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9453
0.0140
0.0121
0.0055
0.0194
0.0036
building
0.0305
0.9532
0.0076
0.0041
0.0018
0.0028
low_veg
0.0401
0.0072
0.8906
0.0579
0.0041
0.0001
tree
0.0391
0.0060
0.0620
0.8924
0.0003
0.0001
car
0.2184
0.1090
0.0168
0.0027
0.6530
0.0000
clutter
0.4060
0.2331
0.0220
0.0107
0.0573
0.2709
Precision/Correctness
0.901
0.933
0.897
0.891
0.689
0.805
Recall/Completeness
0.945
0.953
0.891
0.892
0.653
0.271
F1
0.923
0.943
0.894
0.892
0.670
0.405

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9693
0.0081
0.0077
0.0014
0.0112
0.0023
building
0.0213
0.9694
0.0043
0.0008
0.0015
0.0027
low_veg
0.0251
0.0026
0.9217
0.0486
0.0021
0.0000
tree
0.0134
0.0018
0.0463
0.9384
0.0000
0.0000
car
0.1651
0.1182
0.0128
0.0007
0.7031
0.0000
clutter
0.3621
0.2506
0.0225
0.0041
0.0610
0.2999
Precision/Correctness
0.933
0.950
0.930
0.926
0.744
0.844
Recall/Completeness
0.969
0.969
0.922
0.938
0.703
0.300
F1
0.951
0.960
0.926
0.932
0.723
0.442

Overall accuracy 0.931

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
0.9469
0.0069
0.0141
0.0055
0.0066
0.0200
building
0.0527
0.8817
0.0539
0.0055
0.0015
0.0048
low_veg
0.0547
0.0065
0.8825
0.0539
0.0024
0.0000
tree
0.0350
0.0046
0.1488
0.8116
0.0000
0.0000
car
0.2673
0.0277
0.0074
0.0028
0.6944
0.0005
clutter
0.3467
0.1044
0.0017
0.0056
0.0215
0.5201
Precision/Correctness
0.880
0.962
0.787
0.873
0.664
0.755
Recall/Completeness
0.947
0.882
0.882
0.812
0.694
0.520
F1
0.912
0.920
0.832
0.841
0.679
0.616

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_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9702
0.0028
0.0075
0.0017
0.0035
0.0144
building
0.0453
0.8960
0.0495
0.0033
0.0011
0.0048
low_veg
0.0301
0.0025
0.9161
0.0488
0.0025
0.0000
tree
0.0132
0.0011
0.1394
0.8464
0.0000
0.0000
car
0.1785
0.0339
0.0048
0.0014
0.7811
0.0004
clutter
0.3225
0.1045
0.0005
0.0048
0.0231
0.5446
Precision/Correctness
0.909
0.974
0.812
0.903
0.708
0.796
Recall/Completeness
0.970
0.896
0.916
0.846
0.781
0.545
F1
0.939
0.934
0.861
0.874
0.743
0.647

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8039
0.0523
0.1135
0.0244
0.0059
0.0000
building
0.0271
0.9356
0.0266
0.0098
0.0001
0.0008
low_veg
0.0297
0.0144
0.7008
0.2550
0.0001
0.0000
tree
0.0296
0.0075
0.1033
0.8596
0.0000
0.0000
car
0.2361
0.0097
0.0005
0.0067
0.7470
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.896
0.931
0.744
0.736
0.785
--
Recall/Completeness
0.804
0.936
0.701
0.860
0.747
--
F1
0.848
0.933
0.722
0.793
0.766
--

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_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8554
0.0446
0.0891
0.0079
0.0030
0.0000
building
0.0202
0.9623
0.0143
0.0023
0.0001
0.0009
low_veg
0.0191
0.0059
0.7301
0.2449
0.0000
0.0000
tree
0.0099
0.0026
0.0778
0.9096
0.0000
0.0000
car
0.1344
0.0060
0.0001
0.0012
0.8583
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.940
0.954
0.803
0.769
0.866
--
Recall/Completeness
0.855
0.962
0.730
0.910
0.858
--
F1
0.896
0.958
0.765
0.833
0.862
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8971
0.0367
0.0421
0.0163
0.0078
0.0000
building
0.0432
0.9079
0.0438
0.0041
0.0007
0.0003
low_veg
0.0339
0.0307
0.8465
0.0884
0.0006
0.0000
tree
0.0221
0.0054
0.1308
0.8416
0.0001
0.0000
car
0.2707
0.0335
0.0552
0.0035
0.6370
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.898
0.921
0.830
0.845
0.714
--
Recall/Completeness
0.897
0.908
0.846
0.842
0.637
--
F1
0.898
0.914
0.838
0.843
0.673
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9305
0.0323
0.0272
0.0053
0.0047
0.0000
building
0.0356
0.9323
0.0301
0.0011
0.0006
0.0003
low_veg
0.0199
0.0248
0.8833
0.0717
0.0004
0.0000
tree
0.0080
0.0016
0.1078
0.8825
0.0000
0.0000
car
0.1926
0.0356
0.0467
0.0006
0.7244
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.933
0.939
0.871
0.886
0.775
--
Recall/Completeness
0.930
0.932
0.883
0.882
0.724
--
F1
0.932
0.935
0.877
0.884
0.749
--

Overall accuracy 0.907

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image