Detailled semantic labeling (2D) result


Name B. Yu et al.
Affiliation Chinese Academy of Sciences, China
Abbreviation CAS_Y3
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
0.896
0.047
0.039
0.015
0.003
0.000
building
0.051
0.919
0.026
0.004
0.000
0.000
low_veg
0.051
0.030
0.778
0.141
0.000
0.000
tree
0.022
0.006
0.113
0.859
0.000
0.000
car
0.439
0.039
0.013
0.005
0.504
0.000
clutter
0.451
0.303
0.014
0.008
0.044
0.180
Precision/Correctness
0.865
0.908
0.795
0.844
0.830
0.972
Recall/Completeness
0.896
0.919
0.778
0.859
0.504
0.180
F1
0.880
0.913
0.786
0.852
0.627
0.303

Overall accuracy 0.857

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
0.038
0.031
0.007
0.002
0.000
building
0.044
0.935
0.020
0.001
0.000
0.000
low_veg
0.038
0.019
0.814
0.129
0.000
0.000
tree
0.010
0.002
0.095
0.893
0.000
0.000
car
0.338
0.042
0.010
0.002
0.608
0.000
clutter
0.415
0.323
0.010
0.003
0.047
0.202
Precision/Correctness
0.898
0.932
0.830
0.871
0.861
0.979
Recall/Completeness
0.922
0.935
0.814
0.893
0.608
0.202
F1
0.910
0.933
0.822
0.882
0.713
0.334

Overall accuracy 0.887
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9304
0.0308
0.0238
0.0109
0.0041
0.0000
building
0.0444
0.9314
0.0212
0.0026
0.0001
0.0003
low_veg
0.0668
0.0282
0.7413
0.1637
0.0000
0.0000
tree
0.0317
0.0049
0.0964
0.8660
0.0000
0.0009
car
0.4290
0.0172
0.0017
0.0009
0.5512
0.0000
clutter
0.1037
0.0096
0.0000
0.0256
0.0000
0.8612
Precision/Correctness
0.844
0.929
0.839
0.830
0.892
0.982
Recall/Completeness
0.930
0.931
0.741
0.866
0.551
0.861
F1
0.885
0.930
0.787
0.848
0.681
0.918

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
0.9544
0.0261
0.0138
0.0030
0.0028
0.0000
building
0.0375
0.9478
0.0139
0.0004
0.0001
0.0003
low_veg
0.0496
0.0154
0.7780
0.1571
0.0000
0.0000
tree
0.0159
0.0015
0.0827
0.8995
0.0000
0.0004
car
0.3124
0.0150
0.0010
0.0000
0.6716
0.0000
clutter
0.1045
0.0098
0.0000
0.0081
0.0000
0.8776
Precision/Correctness
0.885
0.954
0.872
0.853
0.920
0.990
Recall/Completeness
0.954
0.948
0.778
0.900
0.672
0.878
F1
0.918
0.951
0.822
0.876
0.776
0.930

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9259
0.0591
0.0088
0.0044
0.0018
0.0000
building
0.0605
0.9323
0.0061
0.0012
0.0000
0.0000
low_veg
0.1338
0.0872
0.5342
0.2449
0.0000
0.0000
tree
0.0558
0.0212
0.1030
0.8200
0.0000
0.0000
car
0.5778
0.0559
0.0017
0.0015
0.3631
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.841
0.958
0.685
0.753
0.906
--
Recall/Completeness
0.926
0.932
0.534
0.820
0.363
--
F1
0.882
0.945
0.600
0.785
0.518
--

Overall accuracy 0.898

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
0.9582
0.0326
0.0062
0.0020
0.0011
0.0000
building
0.0495
0.9448
0.0051
0.0005
0.0000
0.0000
low_veg
0.0983
0.0440
0.5861
0.2716
0.0000
0.0000
tree
0.0267
0.0080
0.0990
0.8662
0.0000
0.0000
car
0.5096
0.0452
0.0010
0.0000
0.4442
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.873
0.979
0.713
0.776
0.939
--
Recall/Completeness
0.958
0.945
0.586
0.866
0.444
--
F1
0.913
0.962
0.643
0.818
0.603
--

Overall accuracy 0.924

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
0.8895
0.0675
0.0302
0.0083
0.0042
0.0003
building
0.0495
0.9363
0.0120
0.0019
0.0004
0.0000
low_veg
0.0815
0.0458
0.7144
0.1579
0.0003
0.0000
tree
0.0155
0.0043
0.0694
0.9107
0.0001
0.0000
car
0.5284
0.0391
0.0011
0.0019
0.4294
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.856
0.928
0.706
0.930
0.847
--
Recall/Completeness
0.889
0.936
0.714
0.911
0.429
--
F1
0.872
0.932
0.710
0.920
0.570
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9243
0.0501
0.0207
0.0016
0.0030
0.0002
building
0.0403
0.9519
0.0070
0.0004
0.0004
0.0000
low_veg
0.0653
0.0272
0.7625
0.1450
0.0000
0.0000
tree
0.0053
0.0010
0.0551
0.9385
0.0000
0.0000
car
0.4346
0.0298
0.0005
0.0001
0.5351
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.895
0.954
0.756
0.950
0.880
--
Recall/Completeness
0.924
0.952
0.763
0.939
0.535
--
F1
0.909
0.953
0.759
0.944
0.665
--

Overall accuracy 0.917

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
0.8949
0.0313
0.0632
0.0080
0.0026
0.0000
building
0.0421
0.9338
0.0206
0.0030
0.0002
0.0004
low_veg
0.0544
0.0196
0.8304
0.0950
0.0007
0.0000
tree
0.0280
0.0077
0.1039
0.8604
0.0000
0.0000
car
0.4404
0.0168
0.0169
0.0038
0.5221
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.890
0.932
0.767
0.904
0.900
--
Recall/Completeness
0.895
0.934
0.830
0.860
0.522
--
F1
0.893
0.933
0.797
0.882
0.661
--

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
0.9219
0.0231
0.0516
0.0021
0.0013
0.0000
building
0.0320
0.9544
0.0123
0.0008
0.0001
0.0004
low_veg
0.0362
0.0112
0.8659
0.0865
0.0003
0.0000
tree
0.0130
0.0040
0.0818
0.9012
0.0000
0.0000
car
0.3398
0.0149
0.0113
0.0027
0.6313
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.929
0.956
0.812
0.925
0.941
--
Recall/Completeness
0.922
0.954
0.866
0.901
0.631
--
F1
0.925
0.955
0.838
0.913
0.756
--

Overall accuracy 0.910

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
0.7995
0.1322
0.0484
0.0173
0.0025
0.0000
building
0.0419
0.9272
0.0211
0.0092
0.0005
0.0000
low_veg
0.0753
0.0406
0.7443
0.1394
0.0003
0.0000
tree
0.0159
0.0086
0.1152
0.8603
0.0000
0.0000
car
0.3832
0.0461
0.0252
0.0016
0.5423
0.0016
clutter
--
--
--
--
--
--
Precision/Correctness
0.911
0.788
0.736
0.842
0.772
--
Recall/Completeness
0.800
0.927
0.744
0.860
0.542
--
F1
0.852
0.852
0.740
0.851
0.637
--

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
0.8299
0.1215
0.0394
0.0076
0.0016
0.0000
building
0.0366
0.9408
0.0155
0.0067
0.0004
0.0000
low_veg
0.0725
0.0255
0.7840
0.1178
0.0003
0.0000
tree
0.0064
0.0040
0.0908
0.8988
0.0000
0.0000
car
0.2973
0.0537
0.0211
0.0001
0.6263
0.0014
clutter
--
--
--
--
--
--
Precision/Correctness
0.930
0.813
0.781
0.880
0.812
--
Recall/Completeness
0.830
0.941
0.784
0.899
0.626
--
F1
0.877
0.872
0.783
0.889
0.707
--

Overall accuracy 0.862

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8769
0.0357
0.0578
0.0286
0.0010
0.0000
building
0.0685
0.9090
0.0186
0.0039
0.0000
0.0000
low_veg
0.0322
0.0334
0.7729
0.1612
0.0003
0.0000
tree
0.0102
0.0082
0.1303
0.8513
0.0000
0.0000
car
0.3911
0.0103
0.0061
0.0023
0.5902
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.816
0.907
0.789
0.859
0.928
--
Recall/Completeness
0.877
0.909
0.773
0.851
0.590
--
F1
0.845
0.908
0.781
0.855
0.721
--

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_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9037
0.0328
0.0492
0.0142
0.0002
0.0000
building
0.0693
0.9194
0.0102
0.0011
0.0000
0.0000
low_veg
0.0289
0.0218
0.8077
0.1414
0.0002
0.0000
tree
0.0048
0.0030
0.1114
0.8808
0.0000
0.0000
car
0.2521
0.0104
0.0037
0.0000
0.7338
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.841
0.937
0.819
0.886
0.965
--
Recall/Completeness
0.904
0.919
0.808
0.881
0.734
--
F1
0.871
0.928
0.814
0.883
0.834
--

Overall accuracy 0.870

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
0.8609
0.0363
0.0689
0.0317
0.0023
0.0000
building
0.0200
0.9534
0.0227
0.0038
0.0001
0.0000
low_veg
0.0756
0.0550
0.7322
0.1370
0.0002
0.0000
tree
0.0265
0.0124
0.1415
0.8196
0.0001
0.0000
car
0.2666
0.0000
0.0005
0.0018
0.7312
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.859
0.864
0.735
0.865
0.761
--
Recall/Completeness
0.861
0.953
0.732
0.820
0.731
--
F1
0.860
0.906
0.734
0.842
0.746
--

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_area14_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8801
0.0318
0.0647
0.0218
0.0016
0.0000
building
0.0152
0.9713
0.0122
0.0013
0.0000
0.0000
low_veg
0.0704
0.0380
0.7649
0.1265
0.0002
0.0000
tree
0.0181
0.0057
0.1277
0.8484
0.0000
0.0000
car
0.1388
0.0000
0.0000
0.0000
0.8612
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.886
0.900
0.759
0.886
0.794
--
Recall/Completeness
0.880
0.971
0.765
0.848
0.861
--
F1
0.883
0.934
0.762
0.867
0.826
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8331
0.0263
0.0890
0.0503
0.0013
0.0000
building
0.0355
0.9286
0.0260
0.0098
0.0000
0.0000
low_veg
0.0167
0.0176
0.8283
0.1361
0.0004
0.0008
tree
0.0264
0.0061
0.1222
0.8453
0.0001
0.0000
car
0.2709
0.0038
0.0448
0.0059
0.6745
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.857
0.935
0.795
0.861
0.811
--
Recall/Completeness
0.833
0.929
0.828
0.845
0.675
--
F1
0.845
0.932
0.811
0.853
0.736
--

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
0.8654
0.0212
0.0811
0.0318
0.0005
0.0000
building
0.0313
0.9477
0.0167
0.0044
0.0000
0.0000
low_veg
0.0126
0.0103
0.8521
0.1238
0.0004
0.0008
tree
0.0185
0.0019
0.1075
0.8720
0.0000
0.0000
car
0.1773
0.0003
0.0383
0.0017
0.7824
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.888
0.961
0.822
0.883
0.852
--
Recall/Completeness
0.865
0.948
0.852
0.872
0.782
--
F1
0.877
0.954
0.837
0.878
0.816
--

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
0.8485
0.0529
0.0514
0.0432
0.0040
0.0000
building
0.0141
0.9518
0.0298
0.0042
0.0000
0.0000
low_veg
0.0353
0.0437
0.7379
0.1831
0.0000
0.0000
tree
0.0070
0.0036
0.0599
0.9295
0.0000
0.0000
car
0.2123
0.0305
0.0631
0.0111
0.6829
0.0000
clutter
0.0011
0.9971
0.0018
0.0000
0.0000
0.0000
Precision/Correctness
0.930
0.766
0.781
0.895
0.777
-nan
Recall/Completeness
0.849
0.952
0.738
0.930
0.683
0.000
F1
0.887
0.849
0.759
0.912
0.727
-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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8839
0.0404
0.0479
0.0251
0.0027
0.0000
building
0.0124
0.9607
0.0254
0.0016
0.0000
0.0000
low_veg
0.0285
0.0298
0.7836
0.1581
0.0000
0.0000
tree
0.0027
0.0010
0.0469
0.9493
0.0000
0.0000
car
0.1221
0.0253
0.0574
0.0037
0.7915
0.0000
clutter
0.0001
0.9997
0.0002
0.0000
0.0000
0.0000
Precision/Correctness
0.953
0.791
0.811
0.920
0.810
-nan
Recall/Completeness
0.884
0.961
0.784
0.949
0.791
0.000
F1
0.917
0.867
0.797
0.934
0.801
-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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8953
0.0302
0.0404
0.0315
0.0026
0.0000
building
0.0521
0.9271
0.0170
0.0036
0.0002
0.0000
low_veg
0.0742
0.0502
0.7506
0.1250
0.0001
0.0000
tree
0.0212
0.0040
0.0968
0.8780
0.0000
0.0000
car
0.4714
0.0108
0.0312
0.0152
0.4715
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.889
0.887
0.774
0.875
0.852
--
Recall/Completeness
0.895
0.927
0.751
0.878
0.471
--
F1
0.892
0.907
0.762
0.877
0.607
--

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_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9267
0.0221
0.0335
0.0161
0.0016
0.0000
building
0.0456
0.9443
0.0089
0.0010
0.0002
0.0000
low_veg
0.0699
0.0339
0.7939
0.1022
0.0001
0.0000
tree
0.0087
0.0012
0.0770
0.9131
0.0000
0.0000
car
0.3818
0.0115
0.0275
0.0068
0.5724
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.916
0.923
0.817
0.912
0.894
--
Recall/Completeness
0.927
0.944
0.794
0.913
0.572
--
F1
0.921
0.934
0.805
0.912
0.698
--

Overall accuracy 0.897

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8132
0.0443
0.1090
0.0308
0.0027
0.0000
building
0.0586
0.8789
0.0557
0.0069
0.0000
0.0000
low_veg
0.0231
0.0194
0.8916
0.0659
0.0001
0.0000
tree
0.0145
0.0078
0.1800
0.7976
0.0000
0.0000
car
0.3408
0.0000
0.0079
0.0031
0.6483
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.823
0.910
0.836
0.860
0.853
--
Recall/Completeness
0.813
0.879
0.892
0.798
0.648
--
F1
0.818
0.894
0.863
0.828
0.736
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8453
0.0334
0.0998
0.0203
0.0012
0.0000
building
0.0564
0.8913
0.0481
0.0042
0.0000
0.0000
low_veg
0.0205
0.0146
0.9089
0.0559
0.0001
0.0000
tree
0.0080
0.0037
0.1562
0.8320
0.0000
0.0000
car
0.2199
0.0000
0.0042
0.0003
0.7756
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.847
0.935
0.859
0.885
0.907
--
Recall/Completeness
0.845
0.891
0.909
0.832
0.776
--
F1
0.846
0.913
0.883
0.858
0.836
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9310
0.0377
0.0206
0.0055
0.0051
0.0000
building
0.0648
0.9114
0.0204
0.0030
0.0004
0.0000
low_veg
0.0685
0.0368
0.7102
0.1842
0.0003
0.0000
tree
0.0140
0.0042
0.0886
0.8932
0.0000
0.0000
car
0.4756
0.0175
0.0055
0.0058
0.4951
0.0004
clutter
0.4334
0.3634
0.1255
0.0046
0.0731
0.0000
Precision/Correctness
0.846
0.938
0.858
0.696
0.889
0.000
Recall/Completeness
0.931
0.911
0.710
0.893
0.495
0.000
F1
0.887
0.924
0.777
0.783
0.636
-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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9563
0.0264
0.0134
0.0019
0.0019
0.0000
building
0.0535
0.9300
0.0150
0.0012
0.0003
0.0000
low_veg
0.0499
0.0208
0.7534
0.1757
0.0002
0.0000
tree
0.0054
0.0010
0.0685
0.9251
0.0000
0.0000
car
0.3802
0.0160
0.0029
0.0046
0.5959
0.0003
clutter
0.4393
0.3851
0.1011
0.0021
0.0723
0.0000
Precision/Correctness
0.885
0.961
0.895
0.724
0.943
0.000
Recall/Completeness
0.956
0.930
0.753
0.925
0.596
0.000
F1
0.919
0.945
0.818
0.812
0.730
-nan

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9271
0.0343
0.0253
0.0111
0.0023
0.0000
building
0.0214
0.9636
0.0108
0.0042
0.0001
0.0000
low_veg
0.0645
0.0242
0.6468
0.2642
0.0004
0.0000
tree
0.0057
0.0021
0.0533
0.9390
0.0000
0.0000
car
0.4038
0.0102
0.0023
0.0086
0.5752
0.0000
clutter
0.5154
0.3611
0.0291
0.0115
0.0801
0.0028
Precision/Correctness
0.915
0.900
0.825
0.694
0.692
1.000
Recall/Completeness
0.927
0.964
0.647
0.939
0.575
0.003
F1
0.921
0.931
0.725
0.798
0.628
0.006

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9485
0.0262
0.0190
0.0051
0.0012
0.0000
building
0.0159
0.9763
0.0060
0.0018
0.0001
0.0000
low_veg
0.0391
0.0097
0.6949
0.2561
0.0001
0.0000
tree
0.0014
0.0004
0.0418
0.9564
0.0000
0.0000
car
0.3021
0.0069
0.0015
0.0054
0.6841
0.0000
clutter
0.4946
0.3861
0.0240
0.0029
0.0890
0.0034
Precision/Correctness
0.939
0.922
0.864
0.729
0.703
1.000
Recall/Completeness
0.948
0.976
0.695
0.956
0.684
0.003
F1
0.944
0.948
0.770
0.827
0.694
0.007

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9673
0.0182
0.0057
0.0039
0.0048
0.0000
building
0.0894
0.8959
0.0103
0.0019
0.0025
0.0000
low_veg
0.0519
0.0023
0.8996
0.0461
0.0001
0.0000
tree
0.0399
0.0055
0.0855
0.8689
0.0003
0.0000
car
0.3217
0.1297
0.0077
0.0055
0.5354
0.0000
clutter
0.6129
0.2504
0.0241
0.0079
0.1040
0.0007
Precision/Correctness
0.845
0.921
0.894
0.913
0.759
0.982
Recall/Completeness
0.967
0.896
0.900
0.869
0.535
0.001
F1
0.902
0.909
0.897
0.891
0.628
0.001

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9787
0.0149
0.0033
0.0006
0.0025
0.0000
building
0.0799
0.9117
0.0059
0.0003
0.0021
0.0000
low_veg
0.0285
0.0004
0.9319
0.0392
0.0000
0.0000
tree
0.0137
0.0016
0.0697
0.9150
0.0000
0.0000
car
0.2167
0.1519
0.0047
0.0010
0.6256
0.0000
clutter
0.5713
0.2876
0.0215
0.0028
0.1160
0.0008
Precision/Correctness
0.879
0.933
0.924
0.941
0.779
0.982
Recall/Completeness
0.979
0.912
0.932
0.915
0.626
0.001
F1
0.926
0.922
0.928
0.928
0.694
0.002

Overall accuracy 0.905

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
0.9660
0.0248
0.0054
0.0035
0.0003
0.0000
building
0.0924
0.8227
0.0820
0.0022
0.0007
0.0000
low_veg
0.0768
0.0355
0.7099
0.1771
0.0007
0.0000
tree
0.0373
0.0072
0.1596
0.7959
0.0000
0.0000
car
0.5654
0.0425
0.0026
0.0025
0.3870
0.0000
clutter
0.7196
0.0980
0.0013
0.0018
0.0208
0.1586
Precision/Correctness
0.808
0.917
0.719
0.740
0.761
1.000
Recall/Completeness
0.966
0.823
0.710
0.796
0.387
0.159
F1
0.880
0.867
0.715
0.767
0.513
0.274

Overall accuracy 0.816

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
0.9778
0.0191
0.0022
0.0009
0.0001
0.0000
building
0.0860
0.8317
0.0810
0.0006
0.0007
0.0000
low_veg
0.0442
0.0257
0.7426
0.1869
0.0006
0.0000
tree
0.0127
0.0024
0.1492
0.8357
0.0000
0.0000
car
0.4675
0.0430
0.0004
0.0008
0.4883
0.0000
clutter
0.6947
0.0968
0.0008
0.0016
0.0242
0.1820
Precision/Correctness
0.837
0.937
0.735
0.748
0.757
1.000
Recall/Completeness
0.978
0.832
0.743
0.836
0.488
0.182
F1
0.902
0.881
0.739
0.789
0.594
0.308

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8031
0.1105
0.0704
0.0155
0.0006
0.0000
building
0.0365
0.9428
0.0159
0.0048
0.0000
0.0000
low_veg
0.0397
0.0262
0.7244
0.2098
0.0000
0.0000
tree
0.0318
0.0067
0.1314
0.8300
0.0001
0.0000
car
0.5023
0.0278
0.0002
0.0098
0.4598
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.867
0.875
0.773
0.772
0.954
--
Recall/Completeness
0.803
0.943
0.724
0.830
0.460
--
F1
0.834
0.907
0.748
0.800
0.621
--

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_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8420
0.1065
0.0476
0.0037
0.0003
0.0000
building
0.0290
0.9622
0.0078
0.0010
0.0000
0.0000
low_veg
0.0259
0.0134
0.7561
0.2046
0.0000
0.0000
tree
0.0107
0.0015
0.1043
0.8835
0.0000
0.0000
car
0.3966
0.0256
0.0000
0.0039
0.5739
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.915
0.900
0.831
0.797
0.981
--
Recall/Completeness
0.842
0.962
0.756
0.884
0.574
--
F1
0.877
0.930
0.792
0.838
0.724
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9053
0.0332
0.0469
0.0123
0.0022
0.0000
building
0.0419
0.8883
0.0641
0.0045
0.0012
0.0000
low_veg
0.0389
0.0201
0.8344
0.1064
0.0002
0.0000
tree
0.0237
0.0036
0.1453
0.8274
0.0001
0.0000
car
0.4597
0.0030
0.0592
0.0093
0.4688
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.888
0.936
0.802
0.825
0.823
--
Recall/Completeness
0.905
0.888
0.834
0.827
0.469
--
F1
0.897
0.912
0.818
0.826
0.597
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9317
0.0300
0.0333
0.0038
0.0012
0.0000
building
0.0351
0.9139
0.0483
0.0017
0.0010
0.0000
low_veg
0.0213
0.0146
0.8693
0.0948
0.0001
0.0000
tree
0.0084
0.0007
0.1208
0.8701
0.0000
0.0000
car
0.3589
0.0031
0.0534
0.0047
0.5799
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.928
0.952
0.842
0.856
0.871
--
Recall/Completeness
0.932
0.914
0.869
0.870
0.580
--
F1
0.930
0.932
0.856
0.863
0.696
--

Overall accuracy 0.896

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image