Detailled semantic labeling (2D) result


Name D. Marmanis et al.
Affiliation DLR-DFD, Germany
Abbreviation DLR_6
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
0.911
0.025
0.046
0.014
0.003
0.000
building
0.082
0.890
0.022
0.006
0.001
0.000
low_veg
0.062
0.020
0.796
0.121
0.000
0.000
tree
0.022
0.006
0.112
0.860
0.000
0.000
car
0.381
0.051
0.011
0.008
0.548
0.000
clutter
0.702
0.242
0.024
0.010
0.022
0.000
Precision/Correctness
0.833
0.937
0.795
0.859
0.856
-nan
Recall/Completeness
0.911
0.890
0.796
0.860
0.548
0.000
F1
0.870
0.913
0.796
0.859
0.668
-nan

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
0.936
0.018
0.038
0.007
0.002
0.000
building
0.074
0.907
0.015
0.004
0.000
0.000
low_veg
0.050
0.012
0.831
0.107
0.000
0.000
tree
0.010
0.002
0.095
0.893
0.000
0.000
car
0.276
0.059
0.008
0.006
0.650
0.000
clutter
0.695
0.256
0.021
0.005
0.023
0.000
Precision/Correctness
0.862
0.958
0.830
0.887
0.886
-nan
Recall/Completeness
0.936
0.907
0.831
0.893
0.650
0.000
F1
0.897
0.932
0.830
0.890
0.750
-nan

Overall accuracy 0.885
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9526
0.0084
0.0258
0.0097
0.0035
0.0000
building
0.0454
0.9278
0.0229
0.0038
0.0001
0.0000
low_veg
0.0838
0.0136
0.7602
0.1424
0.0000
0.0000
tree
0.0328
0.0058
0.0957
0.8656
0.0000
0.0000
car
0.3975
0.0302
0.0019
0.0019
0.5685
0.0000
clutter
0.9819
0.0000
0.0000
0.0181
0.0000
0.0000
Precision/Correctness
0.790
0.967
0.840
0.847
0.907
-nan
Recall/Completeness
0.953
0.928
0.760
0.866
0.569
0.000
F1
0.864
0.947
0.798
0.856
0.699
-nan

Overall accuracy 0.855

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
0.9746
0.0049
0.0150
0.0029
0.0026
0.0000
building
0.0363
0.9492
0.0131
0.0013
0.0001
0.0000
low_veg
0.0656
0.0056
0.7929
0.1359
0.0000
0.0000
tree
0.0164
0.0026
0.0814
0.8996
0.0000
0.0000
car
0.2763
0.0344
0.0010
0.0004
0.6879
0.0000
clutter
0.9967
0.0000
0.0000
0.0033
0.0000
0.0000
Precision/Correctness
0.824
0.984
0.875
0.870
0.926
-nan
Recall/Completeness
0.975
0.949
0.793
0.900
0.688
0.000
F1
0.893
0.966
0.832
0.884
0.790
-nan

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9558
0.0309
0.0076
0.0027
0.0030
0.0000
building
0.0799
0.9052
0.0118
0.0023
0.0008
0.0000
low_veg
0.1518
0.0426
0.5415
0.2636
0.0005
0.0000
tree
0.0597
0.0185
0.1157
0.8061
0.0000
0.0000
car
0.4179
0.0037
0.0017
0.0029
0.5738
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.822
0.977
0.626
0.736
0.860
--
Recall/Completeness
0.956
0.905
0.542
0.806
0.574
--
F1
0.884
0.940
0.581
0.770
0.688
--

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
0.9778
0.0147
0.0051
0.0010
0.0015
0.0000
building
0.0634
0.9237
0.0108
0.0013
0.0008
0.0000
low_veg
0.1054
0.0116
0.5907
0.2919
0.0005
0.0000
tree
0.0261
0.0065
0.1099
0.8576
0.0000
0.0000
car
0.3039
0.0003
0.0001
0.0033
0.6924
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.858
0.991
0.646
0.759
0.896
--
Recall/Completeness
0.978
0.924
0.591
0.858
0.692
--
F1
0.914
0.956
0.617
0.805
0.781
--

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_area6_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8961
0.0422
0.0508
0.0072
0.0037
0.0000
building
0.0728
0.9017
0.0196
0.0048
0.0012
0.0000
low_veg
0.0843
0.0257
0.7441
0.1459
0.0000
0.0000
tree
0.0189
0.0061
0.1350
0.8400
0.0000
0.0000
car
0.4027
0.0042
0.0007
0.0008
0.5915
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.835
0.951
0.576
0.927
0.880
--
Recall/Completeness
0.896
0.902
0.744
0.840
0.592
--
F1
0.865
0.926
0.650
0.881
0.707
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9266
0.0293
0.0404
0.0021
0.0016
0.0000
building
0.0625
0.9196
0.0146
0.0022
0.0011
0.0000
low_veg
0.0708
0.0109
0.7897
0.1286
0.0000
0.0000
tree
0.0077
0.0033
0.1262
0.8629
0.0000
0.0000
car
0.2852
0.0003
0.0002
0.0000
0.7143
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.871
0.971
0.594
0.949
0.920
--
Recall/Completeness
0.927
0.920
0.790
0.863
0.714
--
F1
0.898
0.945
0.678
0.904
0.804
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8919
0.0301
0.0677
0.0084
0.0019
0.0000
building
0.0526
0.9241
0.0181
0.0049
0.0003
0.0000
low_veg
0.0551
0.0115
0.8878
0.0454
0.0002
0.0000
tree
0.0254
0.0074
0.0912
0.8760
0.0000
0.0000
car
0.3944
0.0082
0.0140
0.0040
0.5794
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.887
0.940
0.786
0.943
0.934
--
Recall/Completeness
0.892
0.924
0.888
0.876
0.579
--
F1
0.889
0.932
0.834
0.908
0.715
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9157
0.0248
0.0567
0.0022
0.0006
0.0000
building
0.0406
0.9469
0.0095
0.0027
0.0002
0.0000
low_veg
0.0382
0.0047
0.9190
0.0380
0.0001
0.0000
tree
0.0097
0.0030
0.0688
0.9186
0.0000
0.0000
car
0.2788
0.0067
0.0059
0.0028
0.7058
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.926
0.960
0.830
0.962
0.972
--
Recall/Completeness
0.916
0.947
0.919
0.919
0.706
--
F1
0.921
0.953
0.872
0.940
0.818
--

Overall accuracy 0.922

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
0.8682
0.0399
0.0686
0.0209
0.0024
0.0000
building
0.0352
0.9376
0.0117
0.0150
0.0005
0.0000
low_veg
0.0824
0.0270
0.7210
0.1696
0.0000
0.0000
tree
0.0157
0.0074
0.0794
0.8976
0.0000
0.0000
car
0.3864
0.0157
0.0021
0.0088
0.5870
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.919
0.912
0.739
0.817
0.799
--
Recall/Completeness
0.868
0.938
0.721
0.898
0.587
--
F1
0.893
0.925
0.730
0.856
0.677
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8977
0.0285
0.0623
0.0099
0.0016
0.0000
building
0.0290
0.9525
0.0066
0.0116
0.0004
0.0000
low_veg
0.0815
0.0137
0.7616
0.1431
0.0000
0.0000
tree
0.0054
0.0033
0.0612
0.9301
0.0000
0.0000
car
0.3083
0.0183
0.0000
0.0038
0.6695
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.936
0.943
0.773
0.858
0.836
--
Recall/Completeness
0.898
0.952
0.762
0.930
0.670
--
F1
0.917
0.947
0.767
0.893
0.743
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8874
0.0273
0.0515
0.0322
0.0017
0.0000
building
0.1457
0.8406
0.0085
0.0053
0.0000
0.0000
low_veg
0.0513
0.0273
0.7506
0.1705
0.0003
0.0000
tree
0.0099
0.0073
0.1132
0.8697
0.0000
0.0000
car
0.2616
0.0002
0.0084
0.0070
0.7229
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.726
0.917
0.811
0.854
0.916
--
Recall/Completeness
0.887
0.841
0.751
0.870
0.723
--
F1
0.798
0.877
0.779
0.862
0.808
--

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_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9149
0.0232
0.0434
0.0184
0.0002
0.0000
building
0.1511
0.8445
0.0027
0.0016
0.0000
0.0000
low_veg
0.0489
0.0185
0.7849
0.1476
0.0003
0.0000
tree
0.0042
0.0025
0.0974
0.8959
0.0000
0.0000
car
0.1423
0.0001
0.0061
0.0016
0.8499
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.742
0.945
0.838
0.882
0.970
--
Recall/Completeness
0.915
0.845
0.785
0.896
0.850
--
F1
0.820
0.892
0.811
0.889
0.906
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8442
0.0276
0.0928
0.0304
0.0050
0.0000
building
0.0172
0.9585
0.0196
0.0046
0.0000
0.0000
low_veg
0.0616
0.0452
0.7634
0.1298
0.0000
0.0000
tree
0.0223
0.0133
0.1238
0.8406
0.0000
0.0000
car
0.1779
0.0000
0.0000
0.0331
0.7891
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.879
0.883
0.746
0.873
0.643
--
Recall/Completeness
0.844
0.959
0.763
0.841
0.789
--
F1
0.861
0.919
0.754
0.857
0.708
--

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_area14_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8622
0.0237
0.0883
0.0213
0.0047
0.0000
building
0.0127
0.9755
0.0106
0.0012
0.0000
0.0000
low_veg
0.0564
0.0307
0.7953
0.1177
0.0000
0.0000
tree
0.0144
0.0064
0.1110
0.8682
0.0000
0.0000
car
0.0570
0.0000
0.0000
0.0393
0.9038
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.905
0.917
0.767
0.895
0.621
--
Recall/Completeness
0.862
0.976
0.795
0.868
0.904
--
F1
0.883
0.945
0.781
0.881
0.736
--

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8272
0.0230
0.0984
0.0490
0.0024
0.0000
building
0.0281
0.9460
0.0147
0.0104
0.0008
0.0000
low_veg
0.0248
0.0156
0.8566
0.1029
0.0002
0.0000
tree
0.0203
0.0058
0.1288
0.8451
0.0000
0.0000
car
0.2887
0.0006
0.0024
0.0142
0.6940
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.862
0.942
0.796
0.886
0.732
--
Recall/Completeness
0.827
0.946
0.857
0.845
0.694
--
F1
0.844
0.944
0.825
0.865
0.712
--

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_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8615
0.0184
0.0885
0.0304
0.0012
0.0000
building
0.0246
0.9653
0.0050
0.0043
0.0008
0.0000
low_veg
0.0210
0.0092
0.8818
0.0879
0.0001
0.0000
tree
0.0132
0.0016
0.1142
0.8710
0.0000
0.0000
car
0.2224
0.0000
0.0000
0.0034
0.7742
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.891
0.966
0.823
0.911
0.773
--
Recall/Completeness
0.862
0.965
0.882
0.871
0.774
--
F1
0.876
0.966
0.852
0.890
0.773
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8767
0.0276
0.0548
0.0383
0.0026
0.0000
building
0.0433
0.9236
0.0281
0.0050
0.0000
0.0000
low_veg
0.0386
0.0287
0.7582
0.1745
0.0000
0.0000
tree
0.0077
0.0021
0.0542
0.9359
0.0000
0.0000
car
0.2858
0.0427
0.0150
0.0090
0.6474
0.0000
clutter
0.5494
0.4497
0.0007
0.0002
0.0000
0.0000
Precision/Correctness
0.839
0.861
0.794
0.901
0.836
-nan
Recall/Completeness
0.877
0.924
0.758
0.936
0.647
0.000
F1
0.858
0.891
0.776
0.918
0.730
-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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9077
0.0197
0.0509
0.0202
0.0015
0.0000
building
0.0413
0.9349
0.0225
0.0013
0.0000
0.0000
low_veg
0.0306
0.0189
0.8036
0.1470
0.0000
0.0000
tree
0.0030
0.0006
0.0420
0.9544
0.0000
0.0000
car
0.2029
0.0449
0.0042
0.0013
0.7467
0.0000
clutter
0.5521
0.4477
0.0002
0.0000
0.0000
0.0000
Precision/Correctness
0.859
0.884
0.825
0.927
0.883
-nan
Recall/Completeness
0.908
0.935
0.804
0.954
0.747
0.000
F1
0.882
0.909
0.814
0.940
0.809
-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
0.8913
0.0269
0.0490
0.0277
0.0052
0.0000
building
0.0449
0.9392
0.0097
0.0060
0.0002
0.0000
low_veg
0.0837
0.0340
0.7428
0.1394
0.0001
0.0000
tree
0.0210
0.0046
0.0784
0.8960
0.0000
0.0000
car
0.3325
0.0096
0.0133
0.0066
0.6380
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.892
0.908
0.789
0.871
0.804
--
Recall/Completeness
0.891
0.939
0.743
0.896
0.638
--
F1
0.891
0.924
0.765
0.884
0.711
--

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_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9210
0.0205
0.0418
0.0134
0.0033
0.0000
building
0.0392
0.9549
0.0027
0.0030
0.0002
0.0000
low_veg
0.0797
0.0215
0.7864
0.1124
0.0000
0.0000
tree
0.0080
0.0013
0.0606
0.9302
0.0000
0.0000
car
0.2282
0.0110
0.0132
0.0008
0.7468
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.918
0.939
0.830
0.909
0.847
--
Recall/Completeness
0.921
0.955
0.786
0.930
0.747
--
F1
0.919
0.947
0.807
0.919
0.794
--

Overall accuracy 0.902

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8315
0.0345
0.0948
0.0323
0.0068
0.0000
building
0.0772
0.8769
0.0299
0.0159
0.0000
0.0000
low_veg
0.0500
0.0190
0.8709
0.0601
0.0000
0.0000
tree
0.0188
0.0069
0.1825
0.7919
0.0000
0.0000
car
0.2795
0.0000
0.0020
0.0033
0.7152
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.747
0.919
0.845
0.861
0.729
--
Recall/Completeness
0.832
0.877
0.871
0.792
0.715
--
F1
0.787
0.897
0.858
0.825
0.722
--

Overall accuracy 0.846

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
0.8634
0.0239
0.0855
0.0221
0.0052
0.0000
building
0.0741
0.8897
0.0224
0.0138
0.0000
0.0000
low_veg
0.0472
0.0151
0.8871
0.0507
0.0000
0.0000
tree
0.0125
0.0030
0.1598
0.8247
0.0000
0.0000
car
0.1735
0.0000
0.0000
0.0000
0.8265
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.767
0.942
0.868
0.884
0.748
--
Recall/Completeness
0.863
0.890
0.887
0.825
0.827
--
F1
0.813
0.915
0.877
0.853
0.785
--

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9471
0.0209
0.0219
0.0061
0.0039
0.0000
building
0.0822
0.8956
0.0168
0.0047
0.0006
0.0000
low_veg
0.0840
0.0231
0.7132
0.1797
0.0001
0.0000
tree
0.0125
0.0051
0.0568
0.9256
0.0000
0.0000
car
0.4873
0.0132
0.0069
0.0145
0.4781
0.0000
clutter
0.7971
0.0531
0.0677
0.0023
0.0799
0.0000
Precision/Correctness
0.825
0.962
0.880
0.703
0.903
-nan
Recall/Completeness
0.947
0.896
0.713
0.926
0.478
0.000
F1
0.882
0.928
0.788
0.799
0.625
-nan

Overall accuracy 0.864

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9688
0.0125
0.0153
0.0022
0.0012
0.0000
building
0.0696
0.9169
0.0107
0.0023
0.0005
0.0000
low_veg
0.0642
0.0103
0.7573
0.1681
0.0000
0.0000
tree
0.0050
0.0017
0.0409
0.9524
0.0000
0.0000
car
0.3899
0.0132
0.0055
0.0149
0.5764
0.0000
clutter
0.8140
0.0518
0.0464
0.0000
0.0877
0.0000
Precision/Correctness
0.862
0.981
0.917
0.734
0.950
-nan
Recall/Completeness
0.969
0.917
0.757
0.952
0.576
0.000
F1
0.912
0.948
0.829
0.829
0.717
-nan

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9437
0.0168
0.0250
0.0113
0.0031
0.0000
building
0.0777
0.9048
0.0110
0.0065
0.0000
0.0000
low_veg
0.1050
0.0232
0.6080
0.2636
0.0002
0.0000
tree
0.0061
0.0022
0.0334
0.9582
0.0000
0.0000
car
0.3786
0.0024
0.0064
0.0086
0.6040
0.0000
clutter
0.5519
0.3053
0.0776
0.0160
0.0491
0.0000
Precision/Correctness
0.872
0.925
0.821
0.694
0.743
-nan
Recall/Completeness
0.944
0.905
0.608
0.958
0.604
0.000
F1
0.906
0.915
0.699
0.805
0.666
-nan

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9632
0.0111
0.0182
0.0051
0.0023
0.0000
building
0.0718
0.9187
0.0061
0.0034
0.0000
0.0000
low_veg
0.0860
0.0101
0.6498
0.2540
0.0001
0.0000
tree
0.0018
0.0006
0.0259
0.9718
0.0000
0.0000
car
0.2722
0.0002
0.0069
0.0056
0.7150
0.0000
clutter
0.5389
0.3223
0.0766
0.0073
0.0549
0.0000
Precision/Correctness
0.894
0.945
0.860
0.730
0.745
-nan
Recall/Completeness
0.963
0.919
0.650
0.972
0.715
0.000
F1
0.927
0.932
0.740
0.834
0.730
-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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9622
0.0158
0.0166
0.0033
0.0021
0.0000
building
0.2580
0.7296
0.0091
0.0030
0.0004
0.0000
low_veg
0.0408
0.0032
0.9365
0.0195
0.0000
0.0000
tree
0.0423
0.0040
0.1017
0.8519
0.0001
0.0000
car
0.2828
0.2732
0.0130
0.0091
0.4220
0.0000
clutter
0.5645
0.3717
0.0351
0.0146
0.0142
0.0000
Precision/Correctness
0.771
0.873
0.863
0.943
0.915
-nan
Recall/Completeness
0.962
0.730
0.937
0.852
0.422
0.000
F1
0.856
0.795
0.898
0.895
0.578
-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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9752
0.0107
0.0127
0.0006
0.0009
0.0000
building
0.2520
0.7435
0.0037
0.0005
0.0003
0.0000
low_veg
0.0246
0.0012
0.9604
0.0138
0.0000
0.0000
tree
0.0150
0.0009
0.0823
0.9017
0.0000
0.0000
car
0.1865
0.3049
0.0074
0.0066
0.4945
0.0000
clutter
0.5203
0.4228
0.0289
0.0111
0.0169
0.0000
Precision/Correctness
0.792
0.893
0.899
0.971
0.937
-nan
Recall/Completeness
0.975
0.743
0.960
0.902
0.495
0.000
F1
0.874
0.811
0.929
0.935
0.647
-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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9787
0.0081
0.0105
0.0016
0.0011
0.0000
building
0.2486
0.6528
0.0769
0.0198
0.0020
0.0000
low_veg
0.1025
0.0042
0.8646
0.0277
0.0010
0.0000
tree
0.0516
0.0028
0.1721
0.7735
0.0000
0.0000
car
0.4536
0.0008
0.0042
0.0171
0.5242
0.0000
clutter
0.8264
0.1372
0.0036
0.0043
0.0286
0.0000
Precision/Correctness
0.724
0.943
0.749
0.880
0.712
-nan
Recall/Completeness
0.979
0.653
0.865
0.774
0.524
0.000
F1
0.832
0.771
0.803
0.823
0.604
-nan

Overall accuracy 0.787

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
0.9880
0.0054
0.0057
0.0002
0.0007
0.0000
building
0.2439
0.6612
0.0745
0.0185
0.0019
0.0000
low_veg
0.0707
0.0008
0.9011
0.0263
0.0011
0.0000
tree
0.0230
0.0007
0.1559
0.8203
0.0000
0.0000
car
0.3560
0.0000
0.0012
0.0210
0.6218
0.0000
clutter
0.8145
0.1470
0.0024
0.0045
0.0317
0.0000
Precision/Correctness
0.744
0.954
0.770
0.893
0.705
-nan
Recall/Completeness
0.988
0.661
0.901
0.820
0.622
0.000
F1
0.849
0.781
0.831
0.855
0.661
-nan

Overall accuracy 0.806

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
0.8642
0.0300
0.0926
0.0118
0.0013
0.0000
building
0.0417
0.9333
0.0208
0.0038
0.0004
0.0000
low_veg
0.0390
0.0123
0.7680
0.1806
0.0000
0.0000
tree
0.0305
0.0063
0.1540
0.8092
0.0000
0.0000
car
0.3645
0.0013
0.0004
0.0046
0.6292
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.876
0.954
0.746
0.795
0.918
--
Recall/Completeness
0.864
0.933
0.768
0.809
0.629
--
F1
0.870
0.943
0.757
0.802
0.746
--

Overall accuracy 0.844

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
0.9084
0.0248
0.0633
0.0030
0.0005
0.0000
building
0.0312
0.9586
0.0090
0.0008
0.0004
0.0000
low_veg
0.0253
0.0049
0.7931
0.1767
0.0000
0.0000
tree
0.0102
0.0012
0.1206
0.8680
0.0000
0.0000
car
0.2504
0.0008
0.0000
0.0007
0.7481
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.923
0.973
0.809
0.818
0.951
--
Recall/Completeness
0.908
0.959
0.793
0.868
0.748
--
F1
0.915
0.966
0.801
0.842
0.837
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9041
0.0282
0.0547
0.0104
0.0027
0.0000
building
0.0355
0.9202
0.0393
0.0045
0.0005
0.0000
low_veg
0.0548
0.0129
0.8724
0.0598
0.0000
0.0000
tree
0.0254
0.0041
0.1518
0.8186
0.0000
0.0000
car
0.3682
0.0018
0.0550
0.0042
0.5708
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.881
0.951
0.817
0.885
0.867
--
Recall/Completeness
0.904
0.920
0.872
0.819
0.571
--
F1
0.892
0.935
0.844
0.851
0.688
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9296
0.0255
0.0400
0.0037
0.0013
0.0000
building
0.0268
0.9458
0.0255
0.0015
0.0005
0.0000
low_veg
0.0391
0.0078
0.9052
0.0479
0.0000
0.0000
tree
0.0100
0.0012
0.1233
0.8655
0.0000
0.0000
car
0.2697
0.0002
0.0544
0.0006
0.6751
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.920
0.964
0.858
0.918
0.913
--
Recall/Completeness
0.930
0.946
0.905
0.865
0.675
--
F1
0.925
0.955
0.881
0.891
0.776
--

Overall accuracy 0.913

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image