Detailled semantic labeling (2D) result


Name D. Marmanis et al.
Affiliation DLR-DFD, Germany
Abbreviation DLR_9
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
0.904
0.030
0.045
0.018
0.003
0.000
building
0.036
0.936
0.022
0.006
0.000
0.000
low_veg
0.048
0.022
0.790
0.140
0.000
0.000
tree
0.016
0.005
0.086
0.894
0.000
0.000
car
0.305
0.060
0.007
0.007
0.621
0.000
clutter
0.601
0.361
0.015
0.009
0.015
0.000
Precision/Correctness
0.885
0.931
0.818
0.845
0.889
-nan
Recall/Completeness
0.904
0.936
0.790
0.894
0.621
0.000
F1
0.895
0.933
0.804
0.869
0.731
-nan

Overall accuracy 0.874

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
0.932
0.022
0.036
0.009
0.001
0.000
building
0.028
0.953
0.016
0.003
0.000
0.000
low_veg
0.037
0.014
0.825
0.125
0.000
0.000
tree
0.007
0.002
0.069
0.923
0.000
0.000
car
0.195
0.070
0.005
0.004
0.726
0.000
clutter
0.585
0.386
0.011
0.003
0.016
0.000
Precision/Correctness
0.916
0.950
0.855
0.875
0.921
-nan
Recall/Completeness
0.932
0.953
0.825
0.923
0.726
0.000
F1
0.924
0.952
0.839
0.899
0.812
-nan

Overall accuracy 0.903
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9471
0.0104
0.0262
0.0126
0.0037
0.0000
building
0.0354
0.9406
0.0211
0.0028
0.0000
0.0000
low_veg
0.0734
0.0131
0.7475
0.1661
0.0000
0.0000
tree
0.0261
0.0041
0.0824
0.8873
0.0000
0.0000
car
0.3275
0.0190
0.0008
0.0017
0.6510
0.0000
clutter
0.9798
0.0011
0.0000
0.0190
0.0000
0.0000
Precision/Correctness
0.809
0.968
0.852
0.830
0.915
-nan
Recall/Completeness
0.947
0.941
0.747
0.887
0.651
0.000
F1
0.873
0.954
0.796
0.858
0.761
-nan

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9719
0.0066
0.0154
0.0035
0.0026
0.0000
building
0.0276
0.9600
0.0119
0.0005
0.0000
0.0000
low_veg
0.0571
0.0052
0.7789
0.1588
0.0000
0.0000
tree
0.0121
0.0011
0.0684
0.9183
0.0000
0.0000
car
0.1979
0.0220
0.0003
0.0001
0.7797
0.0000
clutter
0.9954
0.0010
0.0000
0.0036
0.0000
0.0000
Precision/Correctness
0.841
0.985
0.887
0.854
0.935
-nan
Recall/Completeness
0.972
0.960
0.779
0.918
0.780
0.000
F1
0.902
0.972
0.830
0.885
0.850
-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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9388
0.0456
0.0092
0.0037
0.0026
0.0000
building
0.0510
0.9370
0.0094
0.0021
0.0005
0.0000
low_veg
0.1572
0.0375
0.5118
0.2931
0.0005
0.0000
tree
0.0500
0.0140
0.0566
0.8794
0.0000
0.0000
car
0.3496
0.0029
0.0001
0.0003
0.6471
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.864
0.971
0.681
0.733
0.899
--
Recall/Completeness
0.939
0.937
0.512
0.879
0.647
--
F1
0.900
0.954
0.584
0.800
0.752
--

Overall accuracy 0.911

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
0.9662
0.0248
0.0067
0.0011
0.0010
0.0000
building
0.0375
0.9525
0.0083
0.0011
0.0005
0.0000
low_veg
0.1113
0.0085
0.5589
0.3208
0.0005
0.0000
tree
0.0194
0.0050
0.0498
0.9258
0.0000
0.0000
car
0.2274
0.0003
0.0000
0.0000
0.7722
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.902
0.987
0.710
0.759
0.933
--
Recall/Completeness
0.966
0.953
0.559
0.926
0.772
--
F1
0.933
0.969
0.625
0.834
0.845
--

Overall accuracy 0.937

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
0.8941
0.0467
0.0466
0.0100
0.0025
0.0000
building
0.0488
0.9322
0.0152
0.0037
0.0002
0.0000
low_veg
0.0724
0.0232
0.7468
0.1575
0.0000
0.0000
tree
0.0132
0.0032
0.0510
0.9326
0.0000
0.0000
car
0.3854
0.0093
0.0003
0.0010
0.6039
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.870
0.952
0.716
0.929
0.933
--
Recall/Completeness
0.894
0.932
0.747
0.933
0.604
--
F1
0.882
0.942
0.731
0.931
0.733
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9284
0.0320
0.0359
0.0026
0.0010
0.0000
building
0.0388
0.9495
0.0106
0.0009
0.0002
0.0000
low_veg
0.0589
0.0099
0.7949
0.1363
0.0000
0.0000
tree
0.0039
0.0009
0.0396
0.9557
0.0000
0.0000
car
0.2555
0.0075
0.0000
0.0000
0.7370
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.908
0.972
0.761
0.952
0.967
--
Recall/Completeness
0.928
0.950
0.795
0.956
0.737
--
F1
0.918
0.961
0.777
0.954
0.837
--

Overall accuracy 0.928

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
0.8850
0.0323
0.0713
0.0099
0.0016
0.0000
building
0.0313
0.9450
0.0192
0.0044
0.0001
0.0000
low_veg
0.0436
0.0118
0.8946
0.0499
0.0000
0.0000
tree
0.0192
0.0054
0.0699
0.9055
0.0000
0.0000
car
0.3353
0.0162
0.0082
0.0055
0.6347
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.914
0.940
0.801
0.939
0.954
--
Recall/Completeness
0.885
0.945
0.895
0.905
0.635
--
F1
0.899
0.942
0.845
0.922
0.762
--

Overall accuracy 0.901

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
0.9122
0.0261
0.0594
0.0020
0.0002
0.0000
building
0.0213
0.9663
0.0104
0.0019
0.0001
0.0000
low_veg
0.0283
0.0049
0.9283
0.0385
0.0000
0.0000
tree
0.0071
0.0021
0.0485
0.9423
0.0000
0.0000
car
0.2163
0.0180
0.0029
0.0046
0.7583
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.950
0.959
0.846
0.963
0.991
--
Recall/Completeness
0.912
0.966
0.928
0.942
0.758
--
F1
0.931
0.963
0.885
0.953
0.859
--

Overall accuracy 0.934

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
0.8548
0.0642
0.0554
0.0238
0.0018
0.0000
building
0.0214
0.9512
0.0125
0.0149
0.0000
0.0000
low_veg
0.0530
0.0292
0.7482
0.1696
0.0000
0.0000
tree
0.0099
0.0067
0.0639
0.9196
0.0000
0.0000
car
0.3078
0.0310
0.0016
0.0075
0.6521
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.945
0.879
0.781
0.817
0.869
--
Recall/Completeness
0.855
0.951
0.748
0.920
0.652
--
F1
0.898
0.914
0.764
0.865
0.745
--

Overall accuracy 0.871

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
0.8902
0.0512
0.0472
0.0106
0.0008
0.0000
building
0.0170
0.9644
0.0070
0.0117
0.0000
0.0000
low_veg
0.0502
0.0169
0.7935
0.1395
0.0000
0.0000
tree
0.0033
0.0027
0.0475
0.9464
0.0000
0.0000
car
0.2206
0.0371
0.0000
0.0028
0.7394
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.960
0.909
0.820
0.861
0.930
--
Recall/Completeness
0.890
0.964
0.793
0.946
0.739
--
F1
0.924
0.936
0.807
0.902
0.824
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8752
0.0309
0.0541
0.0375
0.0024
0.0000
building
0.0550
0.9308
0.0084
0.0058
0.0000
0.0000
low_veg
0.0368
0.0266
0.7599
0.1767
0.0001
0.0000
tree
0.0071
0.0065
0.0930
0.8934
0.0000
0.0000
car
0.1947
0.0020
0.0094
0.0117
0.7822
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.834
0.925
0.834
0.851
0.916
--
Recall/Completeness
0.875
0.931
0.760
0.893
0.782
--
F1
0.854
0.928
0.795
0.872
0.844
--

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_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9061
0.0273
0.0460
0.0204
0.0002
0.0000
building
0.0563
0.9396
0.0022
0.0019
0.0000
0.0000
low_veg
0.0348
0.0176
0.7955
0.1521
0.0000
0.0000
tree
0.0028
0.0021
0.0777
0.9174
0.0000
0.0000
car
0.0935
0.0025
0.0072
0.0053
0.8915
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.853
0.950
0.863
0.880
0.990
--
Recall/Completeness
0.906
0.940
0.796
0.917
0.892
--
F1
0.879
0.945
0.828
0.898
0.938
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8446
0.0333
0.0800
0.0366
0.0055
0.0000
building
0.0078
0.9654
0.0219
0.0048
0.0000
0.0000
low_veg
0.0512
0.0458
0.7547
0.1482
0.0001
0.0000
tree
0.0194
0.0115
0.1074
0.8617
0.0000
0.0000
car
0.1574
0.0000
0.0000
0.0083
0.8343
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.901
0.880
0.767
0.860
0.634
--
Recall/Completeness
0.845
0.965
0.755
0.862
0.834
--
F1
0.872
0.921
0.761
0.861
0.721
--

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_area14_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8658
0.0289
0.0749
0.0255
0.0049
0.0000
building
0.0041
0.9831
0.0116
0.0012
0.0000
0.0000
low_veg
0.0463
0.0319
0.7869
0.1348
0.0000
0.0000
tree
0.0128
0.0051
0.0950
0.8872
0.0000
0.0000
car
0.0530
0.0000
0.0000
0.0037
0.9434
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.924
0.913
0.791
0.883
0.617
--
Recall/Completeness
0.866
0.983
0.787
0.887
0.943
--
F1
0.894
0.947
0.789
0.885
0.746
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8013
0.0341
0.1057
0.0564
0.0024
0.0000
building
0.0150
0.9561
0.0171
0.0117
0.0000
0.0000
low_veg
0.0136
0.0167
0.8291
0.1404
0.0002
0.0000
tree
0.0153
0.0048
0.0970
0.8829
0.0000
0.0000
car
0.2065
0.0015
0.0005
0.0114
0.7801
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.906
0.934
0.817
0.861
0.799
--
Recall/Completeness
0.801
0.956
0.829
0.883
0.780
--
F1
0.851
0.945
0.823
0.872
0.790
--

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_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8392
0.0294
0.0952
0.0352
0.0011
0.0000
building
0.0126
0.9760
0.0068
0.0046
0.0000
0.0000
low_veg
0.0107
0.0104
0.8543
0.1244
0.0002
0.0000
tree
0.0094
0.0013
0.0834
0.9060
0.0000
0.0000
car
0.1115
0.0000
0.0000
0.0025
0.8861
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.933
0.956
0.846
0.885
0.867
--
Recall/Completeness
0.839
0.976
0.854
0.906
0.886
--
F1
0.884
0.966
0.850
0.896
0.877
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8546
0.0291
0.0623
0.0507
0.0033
0.0000
building
0.0218
0.9421
0.0303
0.0059
0.0000
0.0000
low_veg
0.0299
0.0252
0.7630
0.1819
0.0000
0.0000
tree
0.0052
0.0016
0.0418
0.9514
0.0000
0.0000
car
0.1928
0.0536
0.0242
0.0150
0.7144
0.0000
clutter
0.0108
0.9853
0.0033
0.0005
0.0000
0.0000
Precision/Correctness
0.932
0.801
0.809
0.894
0.817
-nan
Recall/Completeness
0.855
0.942
0.763
0.951
0.714
0.000
F1
0.892
0.866
0.785
0.922
0.762
-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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8916
0.0188
0.0579
0.0298
0.0019
0.0000
building
0.0206
0.9528
0.0251
0.0016
0.0000
0.0000
low_veg
0.0235
0.0150
0.8117
0.1498
0.0000
0.0000
tree
0.0018
0.0003
0.0308
0.9671
0.0000
0.0000
car
0.1092
0.0642
0.0140
0.0058
0.8069
0.0000
clutter
0.0082
0.9906
0.0012
0.0000
0.0000
0.0000
Precision/Correctness
0.952
0.820
0.839
0.922
0.860
-nan
Recall/Completeness
0.892
0.953
0.812
0.967
0.807
0.000
F1
0.921
0.881
0.825
0.944
0.833
-nan

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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8881
0.0289
0.0429
0.0357
0.0044
0.0000
building
0.0154
0.9684
0.0092
0.0068
0.0001
0.0000
low_veg
0.0718
0.0329
0.7441
0.1512
0.0000
0.0000
tree
0.0157
0.0038
0.0511
0.9294
0.0000
0.0000
car
0.2788
0.0244
0.0055
0.0105
0.6808
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.919
0.909
0.833
0.861
0.837
--
Recall/Completeness
0.888
0.968
0.744
0.929
0.681
--
F1
0.903
0.938
0.786
0.894
0.751
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9217
0.0213
0.0365
0.0181
0.0025
0.0000
building
0.0105
0.9832
0.0030
0.0033
0.0000
0.0000
low_veg
0.0683
0.0199
0.7918
0.1201
0.0000
0.0000
tree
0.0050
0.0011
0.0350
0.9589
0.0000
0.0000
car
0.1746
0.0257
0.0031
0.0046
0.7921
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.943
0.941
0.873
0.902
0.888
--
Recall/Completeness
0.922
0.983
0.792
0.959
0.792
--
F1
0.932
0.961
0.831
0.930
0.837
--

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_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8110
0.0409
0.1016
0.0414
0.0051
0.0000
building
0.0577
0.8885
0.0413
0.0125
0.0000
0.0000
low_veg
0.0181
0.0328
0.8688
0.0802
0.0001
0.0000
tree
0.0096
0.0062
0.1466
0.8375
0.0000
0.0000
car
0.2585
0.0000
0.0008
0.0086
0.7321
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.845
0.888
0.857
0.837
0.781
--
Recall/Completeness
0.811
0.889
0.869
0.838
0.732
--
F1
0.827
0.888
0.863
0.837
0.756
--

Overall accuracy 0.856

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
0.8454
0.0293
0.0921
0.0293
0.0038
0.0000
building
0.0570
0.9004
0.0327
0.0100
0.0000
0.0000
low_veg
0.0151
0.0295
0.8862
0.0691
0.0001
0.0000
tree
0.0047
0.0026
0.1241
0.8686
0.0000
0.0000
car
0.1507
0.0000
0.0000
0.0045
0.8448
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.866
0.909
0.880
0.861
0.796
--
Recall/Completeness
0.845
0.900
0.886
0.869
0.845
--
F1
0.855
0.905
0.883
0.865
0.820
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9442
0.0214
0.0229
0.0074
0.0041
0.0000
building
0.0677
0.9106
0.0158
0.0056
0.0004
0.0000
low_veg
0.0903
0.0198
0.6813
0.2085
0.0000
0.0000
tree
0.0103
0.0043
0.0343
0.9511
0.0000
0.0000
car
0.3499
0.0091
0.0035
0.0080
0.6294
0.0000
clutter
0.8088
0.0177
0.1133
0.0036
0.0565
0.0000
Precision/Correctness
0.844
0.965
0.889
0.677
0.928
-nan
Recall/Completeness
0.944
0.911
0.681
0.951
0.629
0.000
F1
0.891
0.937
0.771
0.791
0.750
-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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9678
0.0124
0.0161
0.0027
0.0010
0.0000
building
0.0556
0.9309
0.0104
0.0028
0.0003
0.0000
low_veg
0.0720
0.0077
0.7238
0.1966
0.0000
0.0000
tree
0.0039
0.0011
0.0209
0.9740
0.0000
0.0000
car
0.2366
0.0091
0.0022
0.0063
0.7458
0.0000
clutter
0.8243
0.0085
0.1024
0.0011
0.0637
0.0000
Precision/Correctness
0.881
0.983
0.924
0.708
0.972
-nan
Recall/Completeness
0.968
0.931
0.724
0.974
0.746
0.000
F1
0.922
0.956
0.812
0.820
0.844
-nan

Overall accuracy 0.901

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
0.9390
0.0200
0.0234
0.0140
0.0036
0.0000
building
0.0210
0.9618
0.0103
0.0069
0.0000
0.0000
low_veg
0.0643
0.0209
0.6148
0.2999
0.0001
0.0000
tree
0.0027
0.0016
0.0199
0.9759
0.0000
0.0000
car
0.3018
0.0067
0.0049
0.0103
0.6763
0.0000
clutter
0.5660
0.3601
0.0292
0.0183
0.0264
0.0000
Precision/Correctness
0.916
0.921
0.852
0.669
0.796
-nan
Recall/Completeness
0.939
0.962
0.615
0.976
0.676
0.000
F1
0.928
0.941
0.714
0.794
0.731
-nan

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9601
0.0139
0.0170
0.0063
0.0027
0.0000
building
0.0150
0.9759
0.0056
0.0036
0.0000
0.0000
low_veg
0.0431
0.0086
0.6617
0.2866
0.0000
0.0000
tree
0.0005
0.0003
0.0142
0.9850
0.0000
0.0000
car
0.1990
0.0048
0.0059
0.0072
0.7831
0.0000
clutter
0.5483
0.3875
0.0272
0.0083
0.0287
0.0000
Precision/Correctness
0.939
0.939
0.890
0.708
0.806
-nan
Recall/Completeness
0.960
0.976
0.662
0.985
0.783
0.000
F1
0.949
0.957
0.759
0.824
0.794
-nan

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9589
0.0208
0.0140
0.0040
0.0023
0.0000
building
0.0262
0.9595
0.0106
0.0028
0.0009
0.0000
low_veg
0.0369
0.0054
0.9374
0.0203
0.0000
0.0000
tree
0.0319
0.0029
0.0915
0.8735
0.0001
0.0000
car
0.1978
0.3085
0.0058
0.0086
0.4793
0.0000
clutter
0.5189
0.4328
0.0269
0.0086
0.0128
0.0000
Precision/Correctness
0.901
0.884
0.875
0.943
0.917
-nan
Recall/Completeness
0.959
0.960
0.937
0.874
0.479
0.000
F1
0.929
0.920
0.905
0.907
0.630
-nan

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9732
0.0150
0.0105
0.0005
0.0009
0.0000
building
0.0152
0.9784
0.0051
0.0006
0.0007
0.0000
low_veg
0.0200
0.0032
0.9629
0.0139
0.0000
0.0000
tree
0.0100
0.0005
0.0720
0.9175
0.0000
0.0000
car
0.1081
0.3445
0.0016
0.0044
0.5414
0.0000
clutter
0.4713
0.4893
0.0214
0.0025
0.0154
0.0000
Precision/Correctness
0.936
0.901
0.911
0.974
0.939
-nan
Recall/Completeness
0.973
0.978
0.963
0.918
0.541
0.000
F1
0.954
0.938
0.936
0.945
0.687
-nan

Overall accuracy 0.925

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
0.9725
0.0147
0.0097
0.0023
0.0008
0.0000
building
0.0287
0.8759
0.0815
0.0125
0.0013
0.0000
low_veg
0.0816
0.0105
0.8287
0.0792
0.0000
0.0000
tree
0.0372
0.0032
0.1592
0.8004
0.0000
0.0000
car
0.3910
0.0241
0.0092
0.0066
0.5691
0.0000
clutter
0.7749
0.1999
0.0022
0.0014
0.0216
0.0000
Precision/Correctness
0.839
0.931
0.742
0.833
0.797
-nan
Recall/Completeness
0.973
0.876
0.829
0.800
0.569
0.000
F1
0.901
0.903
0.783
0.816
0.664
-nan

Overall accuracy 0.847

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
0.9846
0.0101
0.0046
0.0003
0.0004
0.0000
building
0.0201
0.8884
0.0797
0.0106
0.0013
0.0000
low_veg
0.0511
0.0052
0.8613
0.0824
0.0000
0.0000
tree
0.0148
0.0004
0.1427
0.8420
0.0000
0.0000
car
0.2871
0.0289
0.0049
0.0076
0.6715
0.0000
clutter
0.7629
0.2114
0.0006
0.0012
0.0239
0.0000
Precision/Correctness
0.871
0.945
0.763
0.844
0.794
-nan
Recall/Completeness
0.985
0.888
0.861
0.842
0.671
0.000
F1
0.924
0.916
0.809
0.843
0.728
-nan

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8580
0.0329
0.0928
0.0156
0.0007
0.0000
building
0.0303
0.9449
0.0196
0.0049
0.0004
0.0000
low_veg
0.0309
0.0095
0.7561
0.2036
0.0000
0.0000
tree
0.0211
0.0046
0.1246
0.8496
0.0000
0.0000
car
0.3488
0.0019
0.0002
0.0072
0.6420
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.902
0.956
0.765
0.781
0.944
--
Recall/Completeness
0.858
0.945
0.756
0.850
0.642
--
F1
0.880
0.950
0.760
0.814
0.764
--

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_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9069
0.0278
0.0620
0.0033
0.0001
0.0000
building
0.0207
0.9701
0.0075
0.0013
0.0004
0.0000
low_veg
0.0193
0.0031
0.7790
0.1987
0.0000
0.0000
tree
0.0055
0.0007
0.0920
0.9018
0.0000
0.0000
car
0.2341
0.0013
0.0000
0.0015
0.7631
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.944
0.972
0.832
0.805
0.973
--
Recall/Completeness
0.907
0.970
0.779
0.902
0.763
--
F1
0.925
0.971
0.805
0.851
0.855
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9016
0.0264
0.0568
0.0134
0.0019
0.0000
building
0.0218
0.9271
0.0447
0.0061
0.0003
0.0000
low_veg
0.0430
0.0161
0.8631
0.0777
0.0000
0.0000
tree
0.0168
0.0028
0.1199
0.8605
0.0000
0.0000
car
0.3003
0.0202
0.0442
0.0076
0.6277
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.911
0.950
0.827
0.862
0.911
--
Recall/Completeness
0.902
0.927
0.863
0.860
0.628
--
F1
0.906
0.938
0.845
0.861
0.743
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9307
0.0237
0.0411
0.0040
0.0005
0.0000
building
0.0141
0.9527
0.0298
0.0030
0.0003
0.0000
low_veg
0.0288
0.0120
0.8960
0.0632
0.0000
0.0000
tree
0.0058
0.0004
0.0932
0.9007
0.0000
0.0000
car
0.1945
0.0228
0.0416
0.0027
0.7384
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.946
0.962
0.870
0.898
0.962
--
Recall/Completeness
0.931
0.953
0.896
0.901
0.738
--
F1
0.938
0.957
0.883
0.899
0.835
--

Overall accuracy 0.920

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image