Detailled semantic labeling (2D) result


Name K. Takahashi
Affiliation Waseda University, Japan
Abbreviation WAS5
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
0.899
0.031
0.049
0.015
0.006
0.001
building
0.060
0.904
0.031
0.004
0.001
0.000
low_veg
0.062
0.019
0.783
0.135
0.000
0.000
tree
0.028
0.006
0.108
0.857
0.000
0.001
car
0.284
0.017
0.018
0.004
0.677
0.000
clutter
0.396
0.311
0.076
0.003
0.057
0.157
Precision/Correctness
0.855
0.932
0.783
0.849
0.790
0.756
Recall/Completeness
0.899
0.904
0.783
0.857
0.677
0.157
F1
0.876
0.917
0.783
0.853
0.729
0.260

Overall accuracy 0.856

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
0.922
0.025
0.042
0.008
0.003
0.001
building
0.051
0.924
0.023
0.002
0.001
0.000
low_veg
0.048
0.012
0.818
0.122
0.000
0.000
tree
0.014
0.002
0.091
0.892
0.000
0.001
car
0.171
0.019
0.016
0.002
0.792
0.000
clutter
0.360
0.331
0.074
0.001
0.061
0.173
Precision/Correctness
0.890
0.951
0.816
0.876
0.822
0.794
Recall/Completeness
0.922
0.924
0.818
0.892
0.792
0.173
F1
0.906
0.937
0.817
0.884
0.807
0.285

Overall accuracy 0.886
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9443
0.0158
0.0232
0.0086
0.0081
0.0000
building
0.0465
0.9262
0.0244
0.0026
0.0003
0.0001
low_veg
0.0892
0.0129
0.7365
0.1609
0.0005
0.0000
tree
0.0369
0.0049
0.0990
0.8541
0.0011
0.0040
car
0.2868
0.0060
0.0024
0.0010
0.7037
0.0000
clutter
0.0000
0.0000
0.0000
0.0022
0.0000
0.9978
Precision/Correctness
0.837
0.962
0.834
0.833
0.815
0.948
Recall/Completeness
0.944
0.926
0.736
0.854
0.704
0.998
F1
0.887
0.944
0.782
0.844
0.755
0.972

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9656
0.0128
0.0133
0.0027
0.0056
0.0000
building
0.0367
0.9501
0.0122
0.0007
0.0003
0.0001
low_veg
0.0707
0.0057
0.7683
0.1549
0.0004
0.0000
tree
0.0192
0.0017
0.0843
0.8915
0.0007
0.0026
car
0.1593
0.0059
0.0024
0.0000
0.8323
0.0000
clutter
0.0000
0.0000
0.0000
0.0002
0.0000
0.9998
Precision/Correctness
0.880
0.978
0.871
0.854
0.851
0.967
Recall/Completeness
0.966
0.950
0.768
0.892
0.832
1.000
F1
0.921
0.964
0.817
0.872
0.842
0.983

Overall accuracy 0.896

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9298
0.0500
0.0095
0.0059
0.0048
0.0000
building
0.0694
0.9196
0.0086
0.0020
0.0005
0.0000
low_veg
0.1899
0.0306
0.4495
0.3292
0.0009
0.0000
tree
0.0606
0.0156
0.0419
0.8817
0.0002
0.0000
car
0.3759
0.0150
0.0016
0.0013
0.6062
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.829
0.968
0.675
0.707
0.837
--
Recall/Completeness
0.930
0.920
0.450
0.882
0.606
--
F1
0.877
0.943
0.540
0.785
0.703
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9530
0.0334
0.0074
0.0032
0.0030
0.0000
building
0.0526
0.9386
0.0073
0.0010
0.0005
0.0000
low_veg
0.1395
0.0085
0.4888
0.3631
0.0002
0.0000
tree
0.0282
0.0067
0.0366
0.9285
0.0000
0.0000
car
0.2603
0.0111
0.0009
0.0001
0.7276
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.870
0.982
0.706
0.731
0.876
--
Recall/Completeness
0.953
0.939
0.489
0.928
0.728
--
F1
0.909
0.960
0.578
0.818
0.795
--

Overall accuracy 0.921

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
0.8965
0.0413
0.0453
0.0070
0.0068
0.0031
building
0.0515
0.9287
0.0168
0.0028
0.0003
0.0000
low_veg
0.1044
0.0286
0.7058
0.1582
0.0003
0.0027
tree
0.0233
0.0070
0.0541
0.9114
0.0003
0.0040
car
0.3514
0.0119
0.0063
0.0016
0.6288
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.849
0.951
0.697
0.930
0.842
--
Recall/Completeness
0.896
0.929
0.706
0.911
0.629
--
F1
0.872
0.940
0.701
0.921
0.720
--

Overall accuracy 0.887

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
0.9268
0.0290
0.0358
0.0024
0.0037
0.0024
building
0.0385
0.9495
0.0109
0.0009
0.0002
0.0000
low_veg
0.0864
0.0185
0.7491
0.1440
0.0000
0.0019
tree
0.0105
0.0026
0.0413
0.9418
0.0001
0.0036
car
0.2364
0.0096
0.0045
0.0006
0.7489
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.893
0.970
0.744
0.949
0.901
--
Recall/Completeness
0.927
0.950
0.749
0.942
0.749
--
F1
0.910
0.960
0.747
0.946
0.818
--

Overall accuracy 0.920

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
0.8694
0.0373
0.0783
0.0081
0.0039
0.0030
building
0.0406
0.9255
0.0314
0.0025
0.0000
0.0000
low_veg
0.0607
0.0083
0.8563
0.0746
0.0001
0.0000
tree
0.0278
0.0060
0.0903
0.8738
0.0001
0.0021
car
0.3178
0.0015
0.0410
0.0038
0.6314
0.0045
clutter
--
--
--
--
--
--
Precision/Correctness
0.891
0.935
0.757
0.921
0.894
--
Recall/Completeness
0.869
0.925
0.856
0.874
0.631
--
F1
0.880
0.930
0.804
0.897
0.740
--

Overall accuracy 0.877

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
0.8924
0.0318
0.0686
0.0022
0.0021
0.0029
building
0.0291
0.9499
0.0206
0.0004
0.0000
0.0000
low_veg
0.0426
0.0029
0.8887
0.0657
0.0000
0.0000
tree
0.0106
0.0024
0.0694
0.9156
0.0000
0.0020
car
0.2097
0.0008
0.0322
0.0028
0.7503
0.0043
clutter
--
--
--
--
--
--
Precision/Correctness
0.932
0.953
0.799
0.942
0.934
--
Recall/Completeness
0.892
0.950
0.889
0.916
0.750
--
F1
0.912
0.952
0.841
0.928
0.832
--

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_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8264
0.0469
0.1039
0.0194
0.0034
0.0000
building
0.0406
0.9281
0.0171
0.0114
0.0026
0.0001
low_veg
0.0572
0.0515
0.7019
0.1889
0.0006
0.0000
tree
0.0180
0.0078
0.0955
0.8787
0.0000
0.0000
car
0.2865
0.0019
0.0001
0.0025
0.7090
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.924
0.886
0.663
0.807
0.714
--
Recall/Completeness
0.826
0.928
0.702
0.879
0.709
--
F1
0.873
0.906
0.682
0.841
0.712
--

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
0.8536
0.0353
0.0994
0.0093
0.0024
0.0000
building
0.0331
0.9442
0.0121
0.0083
0.0022
0.0000
low_veg
0.0526
0.0406
0.7419
0.1643
0.0006
0.0000
tree
0.0074
0.0036
0.0771
0.9119
0.0000
0.0000
car
0.1785
0.0027
0.0000
0.0000
0.8188
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.944
0.915
0.689
0.845
0.732
--
Recall/Completeness
0.854
0.944
0.742
0.912
0.819
--
F1
0.897
0.929
0.715
0.877
0.773
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8933
0.0331
0.0478
0.0228
0.0029
0.0000
building
0.0368
0.9412
0.0185
0.0035
0.0000
0.0000
low_veg
0.0585
0.0267
0.7662
0.1486
0.0000
0.0000
tree
0.0164
0.0082
0.1353
0.8401
0.0001
0.0000
car
0.1824
0.0047
0.0161
0.0023
0.7944
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.804
0.921
0.786
0.869
0.899
--
Recall/Completeness
0.893
0.941
0.766
0.840
0.794
--
F1
0.846
0.931
0.776
0.854
0.844
--

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_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9180
0.0298
0.0391
0.0123
0.0009
0.0000
building
0.0345
0.9556
0.0093
0.0006
0.0000
0.0000
low_veg
0.0544
0.0169
0.8006
0.1280
0.0000
0.0000
tree
0.0089
0.0024
0.1192
0.8695
0.0000
0.0000
car
0.0683
0.0022
0.0140
0.0000
0.9154
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.836
0.949
0.814
0.894
0.964
--
Recall/Completeness
0.918
0.956
0.801
0.870
0.915
--
F1
0.875
0.952
0.807
0.882
0.939
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7911
0.0258
0.1459
0.0320
0.0052
0.0000
building
0.0154
0.9595
0.0199
0.0052
0.0000
0.0000
low_veg
0.0892
0.0448
0.7081
0.1575
0.0004
0.0000
tree
0.0315
0.0143
0.1218
0.8324
0.0000
0.0000
car
0.1640
0.0000
0.0132
0.0078
0.8150
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.833
0.884
0.698
0.852
0.625
--
Recall/Completeness
0.791
0.959
0.708
0.832
0.815
--
F1
0.812
0.920
0.703
0.842
0.707
--

Overall accuracy 0.814

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
0.8056
0.0208
0.1465
0.0225
0.0046
0.0000
building
0.0098
0.9787
0.0097
0.0018
0.0000
0.0000
low_veg
0.0835
0.0304
0.7359
0.1498
0.0003
0.0000
tree
0.0226
0.0072
0.1091
0.8610
0.0000
0.0000
car
0.0616
0.0000
0.0047
0.0039
0.9299
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.861
0.920
0.716
0.872
0.613
--
Recall/Completeness
0.806
0.979
0.736
0.861
0.930
--
F1
0.833
0.948
0.726
0.866
0.739
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8430
0.0319
0.0754
0.0435
0.0030
0.0031
building
0.0406
0.9181
0.0307
0.0099
0.0002
0.0004
low_veg
0.0211
0.0137
0.8672
0.0972
0.0004
0.0004
tree
0.0324
0.0057
0.1315
0.8291
0.0000
0.0013
car
0.2483
0.0011
0.0020
0.0069
0.7417
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.834
0.936
0.797
0.891
0.730
--
Recall/Completeness
0.843
0.918
0.867
0.829
0.742
--
F1
0.839
0.927
0.831
0.859
0.736
--

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
0.8714
0.0270
0.0694
0.0277
0.0018
0.0028
building
0.0354
0.9430
0.0181
0.0032
0.0000
0.0003
low_veg
0.0161
0.0084
0.8915
0.0833
0.0004
0.0003
tree
0.0235
0.0023
0.1179
0.8552
0.0000
0.0012
car
0.1623
0.0000
0.0000
0.0008
0.8369
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.868
0.958
0.823
0.914
0.772
--
Recall/Completeness
0.871
0.943
0.891
0.855
0.837
--
F1
0.870
0.950
0.856
0.884
0.803
--

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
0.8664
0.0457
0.0289
0.0485
0.0094
0.0010
building
0.0191
0.9417
0.0326
0.0065
0.0000
0.0000
low_veg
0.0429
0.0269
0.7327
0.1975
0.0000
0.0000
tree
0.0093
0.0018
0.0538
0.9349
0.0001
0.0000
car
0.1938
0.0000
0.0399
0.0079
0.7584
0.0000
clutter
0.0004
0.9952
0.0041
0.0002
0.0000
0.0000
Precision/Correctness
0.916
0.785
0.809
0.887
0.618
0.000
Recall/Completeness
0.866
0.942
0.733
0.935
0.758
0.000
F1
0.890
0.856
0.769
0.910
0.681
-nan

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8970
0.0356
0.0263
0.0320
0.0082
0.0010
building
0.0162
0.9563
0.0260
0.0015
0.0000
0.0000
low_veg
0.0347
0.0187
0.7788
0.1678
0.0000
0.0000
tree
0.0040
0.0005
0.0431
0.9523
0.0000
0.0000
car
0.0858
0.0000
0.0358
0.0029
0.8756
0.0000
clutter
0.0000
0.9980
0.0020
0.0000
0.0000
0.0000
Precision/Correctness
0.942
0.803
0.838
0.913
0.608
0.000
Recall/Completeness
0.897
0.956
0.779
0.952
0.876
0.000
F1
0.919
0.873
0.807
0.932
0.718
-nan

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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9063
0.0200
0.0360
0.0319
0.0058
0.0001
building
0.0476
0.9272
0.0192
0.0057
0.0004
0.0000
low_veg
0.0868
0.0340
0.7311
0.1482
0.0000
0.0000
tree
0.0258
0.0035
0.0753
0.8951
0.0004
0.0000
car
0.2876
0.0004
0.0150
0.0087
0.6882
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.888
0.921
0.801
0.862
0.790
--
Recall/Completeness
0.906
0.927
0.731
0.895
0.688
--
F1
0.897
0.924
0.764
0.878
0.736
--

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
0.9349
0.0136
0.0296
0.0181
0.0037
0.0001
building
0.0375
0.9493
0.0103
0.0026
0.0004
0.0000
low_veg
0.0805
0.0218
0.7754
0.1224
0.0000
0.0000
tree
0.0110
0.0008
0.0586
0.9294
0.0002
0.0000
car
0.1764
0.0000
0.0091
0.0047
0.8099
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.918
0.951
0.842
0.898
0.832
--
Recall/Completeness
0.935
0.949
0.775
0.929
0.810
--
F1
0.926
0.950
0.807
0.913
0.821
--

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_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8042
0.0383
0.1172
0.0313
0.0083
0.0006
building
0.0981
0.7854
0.1038
0.0125
0.0001
0.0001
low_veg
0.0309
0.0141
0.8706
0.0843
0.0001
0.0000
tree
0.0212
0.0057
0.1679
0.8050
0.0000
0.0001
car
0.2505
0.0000
0.0110
0.0031
0.7354
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.757
0.920
0.819
0.831
0.686
--
Recall/Completeness
0.804
0.785
0.871
0.805
0.735
--
F1
0.780
0.847
0.844
0.818
0.710
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8366
0.0284
0.1082
0.0196
0.0066
0.0005
building
0.0966
0.7933
0.1001
0.0099
0.0001
0.0001
low_veg
0.0281
0.0094
0.8895
0.0729
0.0001
0.0000
tree
0.0143
0.0020
0.1478
0.8358
0.0000
0.0000
car
0.1514
0.0000
0.0046
0.0010
0.8430
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.778
0.947
0.839
0.856
0.696
--
Recall/Completeness
0.837
0.793
0.890
0.836
0.843
--
F1
0.806
0.863
0.863
0.846
0.762
--

Overall accuracy 0.850

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9297
0.0293
0.0247
0.0057
0.0106
0.0000
building
0.0936
0.8823
0.0214
0.0024
0.0002
0.0000
low_veg
0.0815
0.0159
0.7135
0.1890
0.0001
0.0000
tree
0.0152
0.0050
0.0794
0.9002
0.0001
0.0000
car
0.2851
0.0084
0.0090
0.0057
0.6918
0.0000
clutter
0.5328
0.2329
0.1395
0.0002
0.0947
0.0000
Precision/Correctness
0.830
0.957
0.853
0.694
0.858
-nan
Recall/Completeness
0.930
0.882
0.713
0.900
0.692
0.000
F1
0.877
0.918
0.777
0.783
0.766
-nan

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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9539
0.0212
0.0181
0.0020
0.0047
0.0000
building
0.0803
0.9044
0.0144
0.0007
0.0002
0.0000
low_veg
0.0613
0.0060
0.7537
0.1791
0.0000
0.0000
tree
0.0056
0.0019
0.0612
0.9312
0.0000
0.0000
car
0.1703
0.0070
0.0071
0.0039
0.8116
0.0000
clutter
0.5472
0.2256
0.1214
0.0000
0.1059
0.0000
Precision/Correctness
0.867
0.974
0.891
0.723
0.917
-nan
Recall/Completeness
0.954
0.904
0.754
0.931
0.812
0.000
F1
0.908
0.938
0.817
0.814
0.861
-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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9306
0.0249
0.0261
0.0120
0.0064
0.0000
building
0.0307
0.9473
0.0160
0.0059
0.0000
0.0000
low_veg
0.0803
0.0165
0.6277
0.2750
0.0006
0.0000
tree
0.0084
0.0023
0.0531
0.9358
0.0001
0.0004
car
0.2677
0.0026
0.0206
0.0139
0.6953
0.0000
clutter
0.2851
0.5261
0.0548
0.0018
0.1312
0.0010
Precision/Correctness
0.922
0.899
0.801
0.683
0.582
0.394
Recall/Completeness
0.931
0.947
0.628
0.936
0.695
0.001
F1
0.926
0.923
0.704
0.789
0.634
0.002

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9510
0.0187
0.0195
0.0057
0.0050
0.0000
building
0.0239
0.9632
0.0100
0.0029
0.0000
0.0000
low_veg
0.0561
0.0052
0.6723
0.2659
0.0004
0.0000
tree
0.0024
0.0002
0.0441
0.9528
0.0000
0.0004
car
0.1588
0.0003
0.0239
0.0120
0.8050
0.0000
clutter
0.2439
0.5693
0.0423
0.0003
0.1430
0.0011
Precision/Correctness
0.947
0.918
0.842
0.718
0.575
0.380
Recall/Completeness
0.951
0.963
0.672
0.953
0.805
0.001
F1
0.949
0.940
0.748
0.819
0.671
0.002

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9574
0.0136
0.0193
0.0031
0.0066
0.0000
building
0.0702
0.9133
0.0079
0.0046
0.0019
0.0021
low_veg
0.0615
0.0015
0.8912
0.0448
0.0008
0.0003
tree
0.0685
0.0029
0.0557
0.8719
0.0011
0.0000
car
0.2380
0.0732
0.0185
0.0041
0.6659
0.0002
clutter
0.5004
0.1671
0.2544
0.0086
0.0696
0.0000
Precision/Correctness
0.858
0.947
0.840
0.912
0.809
0.000
Recall/Completeness
0.957
0.913
0.891
0.872
0.666
0.000
F1
0.905
0.930
0.865
0.891
0.730
-nan

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9697
0.0105
0.0158
0.0008
0.0032
0.0000
building
0.0597
0.9327
0.0032
0.0007
0.0015
0.0021
low_veg
0.0397
0.0001
0.9231
0.0363
0.0005
0.0003
tree
0.0322
0.0007
0.0435
0.9234
0.0002
0.0000
car
0.1351
0.0862
0.0158
0.0018
0.7608
0.0002
clutter
0.4560
0.1866
0.2764
0.0030
0.0779
0.0000
Precision/Correctness
0.895
0.957
0.870
0.943
0.842
0.000
Recall/Completeness
0.970
0.933
0.923
0.923
0.761
0.000
F1
0.931
0.944
0.896
0.933
0.799
-nan

Overall accuracy 0.912

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
0.9792
0.0073
0.0072
0.0034
0.0029
0.0000
building
0.0934
0.8275
0.0739
0.0039
0.0013
0.0001
low_veg
0.0786
0.0051
0.8537
0.0626
0.0000
0.0000
tree
0.0540
0.0028
0.1588
0.7844
0.0001
0.0000
car
0.2859
0.0002
0.0232
0.0023
0.6883
0.0000
clutter
0.8152
0.0971
0.0294
0.0002
0.0582
0.0000
Precision/Correctness
0.803
0.963
0.755
0.871
0.657
0.000
Recall/Completeness
0.979
0.828
0.854
0.784
0.688
0.000
F1
0.882
0.890
0.801
0.826
0.672
-nan

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_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9892
0.0048
0.0032
0.0014
0.0015
0.0000
building
0.0844
0.8419
0.0702
0.0022
0.0012
0.0001
low_veg
0.0462
0.0028
0.8921
0.0588
0.0000
0.0000
tree
0.0241
0.0006
0.1470
0.8283
0.0000
0.0000
car
0.1548
0.0000
0.0226
0.0013
0.8213
0.0000
clutter
0.8029
0.0986
0.0320
0.0000
0.0665
0.0000
Precision/Correctness
0.833
0.972
0.775
0.894
0.658
0.000
Recall/Completeness
0.989
0.842
0.892
0.828
0.821
0.000
F1
0.904
0.902
0.829
0.860
0.730
-nan

Overall accuracy 0.863

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8241
0.0813
0.0777
0.0128
0.0027
0.0015
building
0.0563
0.9117
0.0267
0.0039
0.0004
0.0010
low_veg
0.0445
0.0097
0.7577
0.1877
0.0003
0.0000
tree
0.0404
0.0066
0.1492
0.8034
0.0001
0.0003
car
0.2341
0.0342
0.0009
0.0030
0.7266
0.0012
clutter
--
--
--
--
--
--
Precision/Correctness
0.847
0.908
0.753
0.787
0.859
--
Recall/Completeness
0.824
0.912
0.758
0.803
0.727
--
F1
0.835
0.910
0.756
0.795
0.787
--

Overall accuracy 0.825

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8599
0.0806
0.0537
0.0033
0.0012
0.0014
building
0.0468
0.9389
0.0124
0.0006
0.0004
0.0009
low_veg
0.0303
0.0033
0.7820
0.1840
0.0004
0.0000
tree
0.0151
0.0021
0.1202
0.8626
0.0000
0.0001
car
0.1223
0.0319
0.0003
0.0001
0.8445
0.0008
clutter
--
--
--
--
--
--
Precision/Correctness
0.895
0.925
0.813
0.811
0.908
--
Recall/Completeness
0.860
0.939
0.782
0.863
0.844
--
F1
0.877
0.932
0.797
0.836
0.875
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9149
0.0216
0.0511
0.0090
0.0033
0.0001
building
0.0832
0.8395
0.0738
0.0024
0.0011
0.0000
low_veg
0.0489
0.0131
0.8632
0.0746
0.0002
0.0000
tree
0.0360
0.0036
0.1575
0.8028
0.0002
0.0000
car
0.3049
0.0015
0.0639
0.0035
0.6260
0.0002
clutter
--
--
--
--
--
--
Precision/Correctness
0.846
0.954
0.791
0.868
0.825
--
Recall/Completeness
0.915
0.840
0.863
0.803
0.626
--
F1
0.879
0.893
0.826
0.834
0.712
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9373
0.0189
0.0393
0.0028
0.0016
0.0001
building
0.0745
0.8659
0.0582
0.0005
0.0010
0.0000
low_veg
0.0301
0.0085
0.9002
0.0611
0.0001
0.0000
tree
0.0158
0.0011
0.1348
0.8482
0.0002
0.0000
car
0.1941
0.0001
0.0632
0.0007
0.7416
0.0002
clutter
--
--
--
--
--
--
Precision/Correctness
0.886
0.968
0.827
0.901
0.879
--
Recall/Completeness
0.937
0.866
0.900
0.848
0.742
--
F1
0.911
0.914
0.862
0.874
0.804
--

Overall accuracy 0.890

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image