Detailled semantic labeling (2D) result


Name D. Marmanis et al. (Plain CNN)
Affiliation DLR-DFD, Germany
Abbreviation DLR_1
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
0.900
0.026
0.048
0.022
0.004
0.000
building
0.080
0.873
0.040
0.006
0.001
0.000
low_veg
0.050
0.024
0.778
0.147
0.000
0.000
tree
0.015
0.005
0.086
0.895
0.000
0.000
car
0.293
0.044
0.009
0.009
0.645
0.000
clutter
0.699
0.254
0.010
0.010
0.027
0.000
Precision/Correctness
0.848
0.932
0.796
0.836
0.837
-nan
Recall/Completeness
0.900
0.873
0.778
0.895
0.645
0.000
F1
0.873
0.902
0.787
0.865
0.728
-nan

Overall accuracy 0.854

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
0.931
0.018
0.038
0.011
0.002
0.000
building
0.072
0.891
0.033
0.003
0.001
0.000
low_veg
0.040
0.016
0.813
0.130
0.000
0.000
tree
0.007
0.002
0.068
0.923
0.000
0.000
car
0.194
0.049
0.005
0.004
0.748
0.000
clutter
0.690
0.272
0.006
0.003
0.028
0.000
Precision/Correctness
0.874
0.954
0.832
0.869
0.872
-nan
Recall/Completeness
0.931
0.891
0.813
0.923
0.748
0.000
F1
0.902
0.921
0.822
0.895
0.805
-nan

Overall accuracy 0.884
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9361
0.0096
0.0303
0.0197
0.0044
0.0000
building
0.0920
0.8735
0.0302
0.0039
0.0005
0.0000
low_veg
0.0681
0.0209
0.7319
0.1789
0.0002
0.0000
tree
0.0218
0.0065
0.0835
0.8882
0.0001
0.0000
car
0.2990
0.0217
0.0003
0.0055
0.6735
0.0000
clutter
0.9697
0.0000
0.0000
0.0303
0.0000
0.0000
Precision/Correctness
0.781
0.955
0.836
0.814
0.893
-nan
Recall/Completeness
0.936
0.873
0.732
0.888
0.674
0.000
F1
0.851
0.912
0.780
0.849
0.768
-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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9679
0.0053
0.0181
0.0061
0.0026
0.0000
building
0.0848
0.8924
0.0214
0.0008
0.0005
0.0000
low_veg
0.0546
0.0117
0.7644
0.1692
0.0001
0.0000
tree
0.0107
0.0025
0.0724
0.9144
0.0000
0.0000
car
0.1720
0.0224
0.0000
0.0018
0.8037
0.0000
clutter
0.9915
0.0000
0.0000
0.0085
0.0000
0.0000
Precision/Correctness
0.806
0.977
0.869
0.843
0.924
-nan
Recall/Completeness
0.968
0.892
0.764
0.914
0.804
0.000
F1
0.880
0.933
0.813
0.877
0.860
-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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9114
0.0654
0.0140
0.0053
0.0040
0.0000
building
0.0840
0.9009
0.0118
0.0022
0.0011
0.0000
low_veg
0.1429
0.0406
0.5136
0.3026
0.0002
0.0000
tree
0.0410
0.0077
0.0614
0.8898
0.0001
0.0000
car
0.4231
0.0163
0.0043
0.0025
0.5537
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.812
0.959
0.625
0.724
0.816
--
Recall/Completeness
0.911
0.901
0.514
0.890
0.554
--
F1
0.859
0.929
0.564
0.798
0.660
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9444
0.0406
0.0107
0.0020
0.0024
0.0000
building
0.0716
0.9159
0.0106
0.0008
0.0012
0.0000
low_veg
0.1050
0.0132
0.5504
0.3311
0.0002
0.0000
tree
0.0146
0.0019
0.0537
0.9298
0.0000
0.0000
car
0.3385
0.0091
0.0018
0.0016
0.6489
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.841
0.978
0.651
0.753
0.839
--
Recall/Completeness
0.944
0.916
0.550
0.930
0.649
--
F1
0.890
0.946
0.596
0.832
0.732
--

Overall accuracy 0.907

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
0.8959
0.0440
0.0421
0.0134
0.0046
0.0000
building
0.0755
0.8767
0.0429
0.0046
0.0003
0.0000
low_veg
0.0665
0.0185
0.7500
0.1644
0.0006
0.0000
tree
0.0089
0.0018
0.0524
0.9367
0.0001
0.0000
car
0.3505
0.0272
0.0008
0.0030
0.6185
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.850
0.954
0.666
0.923
0.880
--
Recall/Completeness
0.896
0.877
0.750
0.937
0.619
--
F1
0.872
0.913
0.706
0.930
0.727
--

Overall accuracy 0.881

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
0.9365
0.0268
0.0305
0.0035
0.0027
0.0000
building
0.0666
0.8955
0.0362
0.0015
0.0002
0.0000
low_veg
0.0535
0.0091
0.7954
0.1416
0.0005
0.0000
tree
0.0022
0.0004
0.0386
0.9588
0.0000
0.0000
car
0.2310
0.0197
0.0003
0.0003
0.7486
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.881
0.975
0.711
0.949
0.920
--
Recall/Completeness
0.937
0.896
0.795
0.959
0.749
--
F1
0.908
0.934
0.751
0.954
0.825
--

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
0.8779
0.0260
0.0793
0.0134
0.0034
0.0000
building
0.0559
0.9076
0.0317
0.0048
0.0001
0.0000
low_veg
0.0382
0.0120
0.8868
0.0626
0.0005
0.0000
tree
0.0171
0.0036
0.0812
0.8980
0.0002
0.0000
car
0.3156
0.0055
0.0144
0.0080
0.6566
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.902
0.948
0.771
0.924
0.902
--
Recall/Completeness
0.878
0.908
0.887
0.898
0.657
--
F1
0.890
0.927
0.825
0.911
0.760
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9096
0.0194
0.0660
0.0035
0.0015
0.0000
building
0.0453
0.9297
0.0229
0.0018
0.0001
0.0000
low_veg
0.0249
0.0050
0.9199
0.0498
0.0003
0.0000
tree
0.0068
0.0011
0.0586
0.9334
0.0002
0.0000
car
0.2094
0.0035
0.0080
0.0054
0.7737
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.934
0.968
0.816
0.952
0.943
--
Recall/Completeness
0.910
0.930
0.920
0.933
0.774
--
F1
0.922
0.948
0.865
0.943
0.850
--

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_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8453
0.0545
0.0733
0.0244
0.0025
0.0000
building
0.0468
0.9105
0.0289
0.0134
0.0005
0.0000
low_veg
0.0586
0.0305
0.7572
0.1533
0.0004
0.0000
tree
0.0090
0.0057
0.0942
0.8911
0.0000
0.0000
car
0.2543
0.0233
0.0049
0.0056
0.7119
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.927
0.888
0.712
0.823
0.819
--
Recall/Completeness
0.845
0.910
0.757
0.891
0.712
--
F1
0.884
0.899
0.734
0.856
0.762
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8814
0.0422
0.0638
0.0113
0.0014
0.0000
building
0.0397
0.9279
0.0226
0.0093
0.0004
0.0000
low_veg
0.0570
0.0193
0.7972
0.1260
0.0005
0.0000
tree
0.0041
0.0025
0.0709
0.9226
0.0000
0.0000
car
0.1599
0.0221
0.0045
0.0009
0.8125
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.942
0.918
0.753
0.869
0.856
--
Recall/Completeness
0.881
0.928
0.797
0.923
0.812
--
F1
0.911
0.923
0.774
0.895
0.834
--

Overall accuracy 0.886

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
0.8743
0.0305
0.0495
0.0416
0.0041
0.0000
building
0.1433
0.8350
0.0128
0.0090
0.0000
0.0000
low_veg
0.0461
0.0314
0.7403
0.1819
0.0003
0.0000
tree
0.0073
0.0071
0.0851
0.9005
0.0000
0.0000
car
0.1962
0.0046
0.0057
0.0098
0.7837
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.737
0.908
0.840
0.846
0.851
--
Recall/Completeness
0.874
0.835
0.740
0.900
0.784
--
F1
0.800
0.870
0.787
0.873
0.816
--

Overall accuracy 0.837

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9073
0.0263
0.0409
0.0241
0.0013
0.0000
building
0.1495
0.8404
0.0072
0.0029
0.0000
0.0000
low_veg
0.0448
0.0188
0.7797
0.1564
0.0003
0.0000
tree
0.0033
0.0028
0.0685
0.9254
0.0000
0.0000
car
0.1068
0.0053
0.0049
0.0011
0.8819
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.749
0.942
0.871
0.877
0.918
--
Recall/Completeness
0.907
0.840
0.780
0.925
0.882
--
F1
0.821
0.888
0.823
0.900
0.899
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8291
0.0284
0.1004
0.0381
0.0040
0.0000
building
0.0272
0.9340
0.0315
0.0071
0.0001
0.0000
low_veg
0.0640
0.0460
0.7387
0.1500
0.0014
0.0000
tree
0.0196
0.0109
0.1106
0.8587
0.0001
0.0000
car
0.1498
0.0000
0.0000
0.0029
0.8473
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.872
0.882
0.741
0.857
0.634
--
Recall/Completeness
0.829
0.934
0.739
0.859
0.847
--
F1
0.850
0.907
0.740
0.858
0.725
--

Overall accuracy 0.835

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
0.8527
0.0237
0.0950
0.0255
0.0031
0.0000
building
0.0229
0.9535
0.0210
0.0025
0.0001
0.0000
low_veg
0.0594
0.0309
0.7714
0.1368
0.0014
0.0000
tree
0.0129
0.0048
0.0978
0.8845
0.0000
0.0000
car
0.0487
0.0000
0.0000
0.0000
0.9513
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.896
0.918
0.765
0.881
0.625
--
Recall/Completeness
0.853
0.953
0.771
0.885
0.951
--
F1
0.874
0.935
0.768
0.883
0.755
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8299
0.0217
0.0842
0.0613
0.0030
0.0000
building
0.0720
0.8736
0.0399
0.0140
0.0004
0.0000
low_veg
0.0237
0.0187
0.8188
0.1385
0.0002
0.0000
tree
0.0209
0.0059
0.0996
0.8736
0.0000
0.0000
car
0.1562
0.0058
0.0170
0.0130
0.8080
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.822
0.933
0.811
0.858
0.750
--
Recall/Completeness
0.830
0.874
0.819
0.874
0.808
--
F1
0.826
0.902
0.815
0.866
0.778
--

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_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8680
0.0158
0.0757
0.0389
0.0016
0.0000
building
0.0697
0.8919
0.0324
0.0056
0.0004
0.0000
low_veg
0.0198
0.0112
0.8451
0.1237
0.0002
0.0000
tree
0.0138
0.0023
0.0860
0.8978
0.0000
0.0000
car
0.0800
0.0018
0.0088
0.0026
0.9067
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.847
0.960
0.837
0.883
0.802
--
Recall/Completeness
0.868
0.892
0.845
0.898
0.907
--
F1
0.858
0.925
0.841
0.890
0.851
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8554
0.0276
0.0551
0.0590
0.0029
0.0000
building
0.0193
0.9329
0.0390
0.0087
0.0001
0.0000
low_veg
0.0531
0.0236
0.7400
0.1829
0.0004
0.0000
tree
0.0054
0.0016
0.0445
0.9484
0.0000
0.0000
car
0.2034
0.0105
0.0188
0.0244
0.7428
0.0000
clutter
0.2822
0.7135
0.0021
0.0021
0.0000
0.0000
Precision/Correctness
0.877
0.835
0.800
0.889
0.820
-nan
Recall/Completeness
0.855
0.933
0.740
0.948
0.743
0.000
F1
0.866
0.881
0.769
0.917
0.779
-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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8903
0.0196
0.0522
0.0365
0.0014
0.0000
building
0.0168
0.9469
0.0330
0.0032
0.0001
0.0000
low_veg
0.0466
0.0146
0.7863
0.1522
0.0004
0.0000
tree
0.0020
0.0003
0.0326
0.9650
0.0000
0.0000
car
0.1188
0.0037
0.0091
0.0159
0.8526
0.0000
clutter
0.2764
0.7221
0.0011
0.0003
0.0000
0.0000
Precision/Correctness
0.897
0.853
0.830
0.918
0.871
-nan
Recall/Completeness
0.890
0.947
0.786
0.965
0.853
0.000
F1
0.894
0.897
0.808
0.941
0.862
-nan

Overall accuracy 0.886

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
0.8820
0.0254
0.0409
0.0471
0.0046
0.0000
building
0.0468
0.9257
0.0166
0.0108
0.0001
0.0000
low_veg
0.0796
0.0394
0.7120
0.1683
0.0006
0.0000
tree
0.0127
0.0031
0.0506
0.9336
0.0000
0.0000
car
0.3398
0.0078
0.0128
0.0230
0.6167
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.899
0.906
0.824
0.840
0.805
--
Recall/Completeness
0.882
0.926
0.712
0.934
0.617
--
F1
0.890
0.916
0.764
0.885
0.698
--

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
0.9207
0.0170
0.0339
0.0256
0.0028
0.0000
building
0.0410
0.9435
0.0101
0.0054
0.0001
0.0000
low_veg
0.0776
0.0252
0.7611
0.1354
0.0007
0.0000
tree
0.0042
0.0006
0.0358
0.9593
0.0000
0.0000
car
0.2526
0.0064
0.0072
0.0116
0.7222
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.920
0.941
0.865
0.885
0.847
--
Recall/Completeness
0.921
0.943
0.761
0.959
0.722
--
F1
0.920
0.942
0.810
0.921
0.780
--

Overall accuracy 0.903

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7975
0.0335
0.1179
0.0454
0.0057
0.0000
building
0.0601
0.8622
0.0644
0.0133
0.0000
0.0000
low_veg
0.0159
0.0252
0.8802
0.0783
0.0005
0.0000
tree
0.0125
0.0066
0.1402
0.8408
0.0000
0.0000
car
0.2406
0.0021
0.0080
0.0099
0.7394
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.841
0.906
0.848
0.838
0.731
--
Recall/Completeness
0.797
0.862
0.880
0.841
0.739
--
F1
0.819
0.883
0.864
0.839
0.735
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8331
0.0226
0.1079
0.0324
0.0040
0.0000
building
0.0581
0.8727
0.0584
0.0108
0.0000
0.0000
low_veg
0.0132
0.0205
0.8992
0.0666
0.0005
0.0000
tree
0.0078
0.0029
0.1170
0.8723
0.0000
0.0000
car
0.1504
0.0000
0.0008
0.0039
0.8449
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.863
0.930
0.871
0.863
0.744
--
Recall/Completeness
0.833
0.873
0.899
0.872
0.845
--
F1
0.848
0.901
0.885
0.868
0.791
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9342
0.0240
0.0257
0.0088
0.0073
0.0000
building
0.1148
0.8479
0.0312
0.0052
0.0009
0.0000
low_veg
0.0777
0.0193
0.6801
0.2226
0.0004
0.0000
tree
0.0092
0.0023
0.0363
0.9522
0.0000
0.0000
car
0.2994
0.0198
0.0060
0.0091
0.6656
0.0000
clutter
0.7488
0.1071
0.0931
0.0086
0.0424
0.0000
Precision/Correctness
0.815
0.959
0.852
0.663
0.885
-nan
Recall/Completeness
0.934
0.848
0.680
0.952
0.666
0.000
F1
0.870
0.900
0.756
0.782
0.760
-nan

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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9628
0.0152
0.0165
0.0034
0.0021
0.0000
building
0.1030
0.8693
0.0247
0.0022
0.0008
0.0000
low_veg
0.0628
0.0084
0.7182
0.2104
0.0002
0.0000
tree
0.0036
0.0005
0.0217
0.9742
0.0000
0.0000
car
0.2002
0.0199
0.0034
0.0067
0.7698
0.0000
clutter
0.7761
0.1002
0.0677
0.0063
0.0497
0.0000
Precision/Correctness
0.845
0.978
0.891
0.694
0.945
-nan
Recall/Completeness
0.963
0.869
0.718
0.974
0.770
0.000
F1
0.900
0.920
0.795
0.811
0.849
-nan

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9387
0.0161
0.0249
0.0164
0.0038
0.0000
building
0.0526
0.9277
0.0118
0.0078
0.0001
0.0000
low_veg
0.0489
0.0188
0.6208
0.3113
0.0002
0.0000
tree
0.0030
0.0014
0.0222
0.9733
0.0000
0.0000
car
0.3093
0.0048
0.0008
0.0129
0.6722
0.0000
clutter
0.5516
0.3695
0.0320
0.0105
0.0364
0.0000
Precision/Correctness
0.904
0.924
0.843
0.657
0.765
-nan
Recall/Completeness
0.939
0.928
0.621
0.973
0.672
0.000
F1
0.921
0.926
0.715
0.784
0.716
-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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9625
0.0097
0.0174
0.0078
0.0026
0.0000
building
0.0466
0.9420
0.0069
0.0044
0.0001
0.0000
low_veg
0.0309
0.0077
0.6625
0.2990
0.0000
0.0000
tree
0.0007
0.0002
0.0165
0.9826
0.0000
0.0000
car
0.2118
0.0045
0.0006
0.0101
0.7730
0.0000
clutter
0.5321
0.4012
0.0224
0.0046
0.0397
0.0000
Precision/Correctness
0.925
0.943
0.885
0.696
0.779
-nan
Recall/Completeness
0.962
0.942
0.662
0.983
0.773
0.000
F1
0.943
0.942
0.758
0.815
0.776
-nan

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
0.9669
0.0094
0.0145
0.0049
0.0042
0.0000
building
0.1488
0.8365
0.0099
0.0034
0.0014
0.0000
low_veg
0.0594
0.0037
0.8958
0.0409
0.0002
0.0000
tree
0.0315
0.0019
0.0649
0.9012
0.0004
0.0000
car
0.2273
0.1908
0.0055
0.0084
0.5680
0.0000
clutter
0.7162
0.2342
0.0042
0.0129
0.0325
0.0000
Precision/Correctness
0.814
0.924
0.892
0.914
0.865
-nan
Recall/Completeness
0.967
0.837
0.896
0.901
0.568
0.000
F1
0.884
0.878
0.894
0.907
0.686
-nan

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9798
0.0055
0.0120
0.0011
0.0017
0.0000
building
0.1377
0.8543
0.0059
0.0008
0.0013
0.0000
low_veg
0.0362
0.0017
0.9290
0.0331
0.0000
0.0000
tree
0.0102
0.0002
0.0483
0.9413
0.0000
0.0000
car
0.1436
0.2127
0.0018
0.0026
0.6393
0.0000
clutter
0.6906
0.2655
0.0028
0.0041
0.0370
0.0000
Precision/Correctness
0.842
0.938
0.922
0.947
0.894
-nan
Recall/Completeness
0.980
0.854
0.929
0.941
0.639
0.000
F1
0.906
0.894
0.925
0.944
0.746
-nan

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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9790
0.0082
0.0084
0.0032
0.0012
0.0000
building
0.1138
0.7113
0.1671
0.0045
0.0033
0.0000
low_veg
0.0776
0.0046
0.8505
0.0667
0.0005
0.0000
tree
0.0392
0.0012
0.1404
0.8193
0.0000
0.0000
car
0.3424
0.0260
0.0069
0.0006
0.6241
0.0000
clutter
0.8509
0.1014
0.0038
0.0003
0.0437
0.0000
Precision/Correctness
0.791
0.955
0.663
0.870
0.675
-nan
Recall/Completeness
0.979
0.711
0.851
0.819
0.624
0.000
F1
0.875
0.815
0.745
0.844
0.649
-nan

Overall accuracy 0.808

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
0.9910
0.0043
0.0035
0.0008
0.0004
0.0000
building
0.1054
0.7221
0.1676
0.0019
0.0031
0.0000
low_veg
0.0458
0.0024
0.8867
0.0649
0.0003
0.0000
tree
0.0146
0.0001
0.1263
0.8591
0.0000
0.0000
car
0.2245
0.0284
0.0038
0.0000
0.7433
0.0000
clutter
0.8451
0.1053
0.0017
0.0000
0.0479
0.0000
Precision/Correctness
0.818
0.966
0.674
0.892
0.676
-nan
Recall/Completeness
0.991
0.722
0.887
0.859
0.743
0.000
F1
0.896
0.827
0.766
0.875
0.708
-nan

Overall accuracy 0.829

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
0.8468
0.0343
0.0910
0.0263
0.0016
0.0000
building
0.0838
0.8825
0.0255
0.0070
0.0011
0.0000
low_veg
0.0355
0.0108
0.7187
0.2346
0.0004
0.0000
tree
0.0175
0.0038
0.1137
0.8650
0.0000
0.0000
car
0.3080
0.0030
0.0058
0.0165
0.6668
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.850
0.951
0.760
0.752
0.861
--
Recall/Completeness
0.847
0.882
0.719
0.865
0.667
--
F1
0.848
0.915
0.739
0.805
0.751
--

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_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9025
0.0271
0.0629
0.0072
0.0004
0.0000
building
0.0728
0.9096
0.0143
0.0021
0.0011
0.0000
low_veg
0.0250
0.0029
0.7490
0.2225
0.0005
0.0000
tree
0.0053
0.0010
0.0848
0.9089
0.0000
0.0000
car
0.2025
0.0019
0.0023
0.0074
0.7858
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.884
0.971
0.825
0.785
0.899
--
Recall/Completeness
0.903
0.910
0.749
0.909
0.786
--
F1
0.893
0.939
0.785
0.842
0.838
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9100
0.0159
0.0548
0.0161
0.0032
0.0000
building
0.0558
0.8524
0.0854
0.0052
0.0012
0.0000
low_veg
0.0399
0.0314
0.8434
0.0849
0.0004
0.0000
tree
0.0138
0.0017
0.1174
0.8669
0.0001
0.0000
car
0.2607
0.0278
0.0443
0.0156
0.6515
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.889
0.941
0.798
0.851
0.828
--
Recall/Completeness
0.910
0.852
0.843
0.867
0.652
--
F1
0.899
0.894
0.820
0.859
0.729
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9437
0.0126
0.0372
0.0048
0.0017
0.0000
building
0.0488
0.8770
0.0713
0.0018
0.0011
0.0000
low_veg
0.0281
0.0293
0.8744
0.0678
0.0004
0.0000
tree
0.0044
0.0002
0.0910
0.9043
0.0001
0.0000
car
0.1715
0.0311
0.0351
0.0058
0.7566
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.918
0.952
0.839
0.893
0.864
--
Recall/Completeness
0.944
0.877
0.874
0.904
0.757
--
F1
0.930
0.913
0.856
0.899
0.807
--

Overall accuracy 0.898

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image