Detailled semantic labeling (2D) result


Name G. Chen and X. Zhang
Affiliation Wuhan University, China
Abbreviation CVEO
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
0.912
0.033
0.039
0.013
0.003
0.000
building
0.075
0.893
0.028
0.004
0.001
0.000
low_veg
0.064
0.026
0.771
0.139
0.000
0.000
tree
0.025
0.005
0.108
0.862
0.000
0.000
car
0.296
0.067
0.011
0.004
0.622
0.000
clutter
0.367
0.352
0.020
0.006
0.041
0.213
Precision/Correctness
0.846
0.921
0.796
0.848
0.857
0.985
Recall/Completeness
0.912
0.893
0.771
0.862
0.622
0.213
F1
0.878
0.907
0.783
0.855
0.721
0.351

Overall accuracy 0.855

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
0.935
0.026
0.031
0.006
0.002
0.000
building
0.068
0.909
0.021
0.002
0.000
0.000
low_veg
0.051
0.019
0.803
0.127
0.000
0.000
tree
0.012
0.002
0.090
0.896
0.000
0.000
car
0.194
0.076
0.006
0.001
0.722
0.000
clutter
0.328
0.375
0.017
0.002
0.041
0.237
Precision/Correctness
0.877
0.940
0.831
0.874
0.882
0.991
Recall/Completeness
0.935
0.909
0.803
0.896
0.722
0.237
F1
0.905
0.924
0.817
0.885
0.794
0.383

Overall accuracy 0.883
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9532
0.0130
0.0208
0.0094
0.0035
0.0000
building
0.0643
0.9106
0.0223
0.0025
0.0003
0.0000
low_veg
0.0890
0.0179
0.7137
0.1790
0.0004
0.0000
tree
0.0342
0.0027
0.1012
0.8613
0.0000
0.0005
car
0.3593
0.0197
0.0016
0.0006
0.6188
0.0000
clutter
0.0668
0.0000
0.0035
0.0140
0.0000
0.9157
Precision/Correctness
0.822
0.960
0.831
0.819
0.902
0.993
Recall/Completeness
0.953
0.911
0.714
0.861
0.619
0.916
F1
0.883
0.935
0.768
0.840
0.734
0.953

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9729
0.0102
0.0120
0.0022
0.0028
0.0000
building
0.0550
0.9319
0.0121
0.0008
0.0003
0.0000
low_veg
0.0716
0.0113
0.7419
0.1749
0.0004
0.0000
tree
0.0173
0.0006
0.0860
0.8959
0.0000
0.0001
car
0.2313
0.0228
0.0006
0.0000
0.7453
0.0000
clutter
0.0604
0.0000
0.0024
0.0025
0.0000
0.9347
Precision/Correctness
0.862
0.975
0.867
0.840
0.915
0.998
Recall/Completeness
0.973
0.932
0.742
0.896
0.745
0.935
F1
0.914
0.953
0.799
0.867
0.821
0.965

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9162
0.0633
0.0134
0.0042
0.0029
0.0000
building
0.0848
0.9017
0.0105
0.0021
0.0008
0.0000
low_veg
0.1628
0.0363
0.5231
0.2762
0.0016
0.0000
tree
0.0607
0.0121
0.0994
0.8276
0.0003
0.0000
car
0.4548
0.0583
0.0046
0.0028
0.4795
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.806
0.959
0.614
0.729
0.833
--
Recall/Completeness
0.916
0.902
0.523
0.828
0.480
--
F1
0.858
0.930
0.565
0.775
0.609
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9426
0.0432
0.0103
0.0018
0.0021
0.0001
building
0.0724
0.9163
0.0092
0.0012
0.0008
0.0000
low_veg
0.1225
0.0098
0.5587
0.3074
0.0017
0.0000
tree
0.0288
0.0042
0.0930
0.8739
0.0001
0.0000
car
0.3730
0.0558
0.0019
0.0017
0.5676
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.835
0.976
0.637
0.752
0.846
--
Recall/Completeness
0.943
0.916
0.559
0.874
0.568
--
F1
0.886
0.945
0.595
0.808
0.679
--

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_area6_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8992
0.0623
0.0299
0.0066
0.0020
0.0000
building
0.0826
0.8887
0.0263
0.0023
0.0002
0.0000
low_veg
0.0770
0.0283
0.7575
0.1368
0.0004
0.0000
tree
0.0178
0.0032
0.0754
0.9036
0.0001
0.0000
car
0.3669
0.0756
0.0069
0.0009
0.5489
0.0009
clutter
--
--
--
--
--
--
Precision/Correctness
0.831
0.933
0.679
0.939
0.935
--
Recall/Completeness
0.899
0.889
0.757
0.904
0.549
--
F1
0.864
0.910
0.716
0.921
0.692
--

Overall accuracy 0.876

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
0.9322
0.0449
0.0203
0.0018
0.0008
0.0000
building
0.0756
0.9025
0.0211
0.0007
0.0001
0.0000
low_veg
0.0581
0.0155
0.8038
0.1225
0.0002
0.0000
tree
0.0065
0.0008
0.0592
0.9335
0.0000
0.0000
car
0.2604
0.0721
0.0043
0.0000
0.6621
0.0011
clutter
--
--
--
--
--
--
Precision/Correctness
0.864
0.957
0.723
0.956
0.971
--
Recall/Completeness
0.932
0.903
0.804
0.933
0.662
--
F1
0.897
0.929
0.761
0.945
0.787
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8954
0.0293
0.0673
0.0058
0.0022
0.0000
building
0.0612
0.9169
0.0179
0.0039
0.0002
0.0000
low_veg
0.0527
0.0141
0.8830
0.0499
0.0001
0.0000
tree
0.0266
0.0057
0.1105
0.8571
0.0000
0.0000
car
0.3347
0.0204
0.0255
0.0011
0.6182
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.885
0.939
0.769
0.942
0.932
--
Recall/Completeness
0.895
0.917
0.883
0.857
0.618
--
F1
0.890
0.928
0.822
0.898
0.743
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9181
0.0237
0.0557
0.0012
0.0012
0.0000
building
0.0529
0.9349
0.0101
0.0019
0.0001
0.0000
low_veg
0.0378
0.0064
0.9153
0.0405
0.0000
0.0000
tree
0.0096
0.0026
0.0872
0.9006
0.0000
0.0000
car
0.2197
0.0213
0.0177
0.0004
0.7409
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.920
0.959
0.813
0.961
0.958
--
Recall/Completeness
0.918
0.935
0.915
0.901
0.741
--
F1
0.919
0.947
0.861
0.930
0.835
--

Overall accuracy 0.915

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
0.8458
0.0794
0.0571
0.0157
0.0020
0.0000
building
0.0420
0.9303
0.0159
0.0113
0.0006
0.0000
low_veg
0.0800
0.0324
0.7490
0.1383
0.0004
0.0000
tree
0.0163
0.0070
0.0842
0.8925
0.0000
0.0000
car
0.2125
0.0414
0.0093
0.0061
0.7307
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.917
0.855
0.756
0.848
0.846
--
Recall/Completeness
0.846
0.930
0.749
0.892
0.731
--
F1
0.880
0.891
0.752
0.870
0.784
--

Overall accuracy 0.858

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8750
0.0700
0.0485
0.0057
0.0009
0.0000
building
0.0366
0.9430
0.0113
0.0087
0.0004
0.0000
low_veg
0.0759
0.0209
0.7903
0.1125
0.0004
0.0000
tree
0.0068
0.0034
0.0640
0.9258
0.0000
0.0000
car
0.1122
0.0466
0.0073
0.0017
0.8323
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.934
0.879
0.796
0.888
0.897
--
Recall/Completeness
0.875
0.943
0.790
0.926
0.832
--
F1
0.904
0.910
0.793
0.907
0.863
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8865
0.0336
0.0528
0.0249
0.0022
0.0000
building
0.1409
0.8428
0.0125
0.0037
0.0000
0.0001
low_veg
0.0585
0.0238
0.7702
0.1473
0.0002
0.0000
tree
0.0121
0.0072
0.1236
0.8571
0.0000
0.0000
car
0.2026
0.0006
0.0098
0.0050
0.7821
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.718
0.918
0.800
0.871
0.910
--
Recall/Completeness
0.887
0.843
0.770
0.857
0.782
--
F1
0.794
0.879
0.785
0.864
0.841
--

Overall accuracy 0.833

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
0.9120
0.0317
0.0437
0.0121
0.0005
0.0000
building
0.1473
0.8452
0.0063
0.0011
0.0000
0.0001
low_veg
0.0560
0.0154
0.8029
0.1255
0.0001
0.0000
tree
0.0061
0.0029
0.1068
0.8842
0.0000
0.0000
car
0.0921
0.0004
0.0095
0.0002
0.8978
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.734
0.943
0.828
0.898
0.967
--
Recall/Completeness
0.912
0.845
0.803
0.884
0.898
--
F1
0.814
0.891
0.815
0.891
0.931
--

Overall accuracy 0.858

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8445
0.0335
0.0873
0.0308
0.0039
0.0000
building
0.0160
0.9592
0.0203
0.0044
0.0000
0.0000
low_veg
0.0867
0.0496
0.7054
0.1580
0.0004
0.0000
tree
0.0336
0.0103
0.1210
0.8350
0.0000
0.0000
car
0.1640
0.0021
0.0008
0.0039
0.8292
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.841
0.876
0.737
0.853
0.685
--
Recall/Completeness
0.845
0.959
0.705
0.835
0.829
--
F1
0.843
0.916
0.721
0.844
0.750
--

Overall accuracy 0.826

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
0.8639
0.0287
0.0835
0.0207
0.0032
0.0000
building
0.0109
0.9764
0.0112
0.0015
0.0000
0.0000
low_veg
0.0813
0.0364
0.7334
0.1485
0.0004
0.0000
tree
0.0249
0.0042
0.1081
0.8627
0.0000
0.0000
car
0.0578
0.0031
0.0002
0.0002
0.9387
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.868
0.908
0.759
0.874
0.682
--
Recall/Completeness
0.864
0.976
0.733
0.863
0.939
--
F1
0.866
0.941
0.746
0.868
0.790
--

Overall accuracy 0.851

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
0.8385
0.0328
0.0797
0.0465
0.0025
0.0000
building
0.0458
0.9027
0.0437
0.0075
0.0003
0.0000
low_veg
0.0248
0.0194
0.8205
0.1352
0.0001
0.0000
tree
0.0255
0.0041
0.1367
0.8336
0.0000
0.0000
car
0.1824
0.0000
0.0101
0.0024
0.8051
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.837
0.929
0.776
0.863
0.795
--
Recall/Completeness
0.838
0.903
0.820
0.834
0.805
--
F1
0.838
0.915
0.798
0.848
0.800
--

Overall accuracy 0.843

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
0.8718
0.0293
0.0687
0.0292
0.0011
0.0000
building
0.0426
0.9219
0.0327
0.0025
0.0002
0.0000
low_veg
0.0208
0.0142
0.8425
0.1225
0.0000
0.0000
tree
0.0176
0.0012
0.1215
0.8597
0.0000
0.0000
car
0.1099
0.0000
0.0044
0.0000
0.8856
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.865
0.948
0.804
0.885
0.867
--
Recall/Completeness
0.872
0.922
0.842
0.860
0.886
--
F1
0.869
0.935
0.823
0.872
0.876
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8479
0.0595
0.0555
0.0343
0.0028
0.0000
building
0.0262
0.9435
0.0247
0.0055
0.0000
0.0000
low_veg
0.0603
0.0279
0.7614
0.1504
0.0001
0.0000
tree
0.0085
0.0023
0.0595
0.9298
0.0000
0.0000
car
0.2100
0.0170
0.0198
0.0123
0.7409
0.0000
clutter
0.0137
0.9848
0.0012
0.0003
0.0000
0.0000
Precision/Correctness
0.892
0.774
0.788
0.912
0.841
0.000
Recall/Completeness
0.848
0.944
0.761
0.930
0.741
0.000
F1
0.869
0.851
0.774
0.921
0.788
-nan

Overall accuracy 0.857

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8782
0.0505
0.0508
0.0191
0.0014
0.0000
building
0.0244
0.9546
0.0190
0.0021
0.0000
0.0000
low_veg
0.0545
0.0164
0.8039
0.1251
0.0001
0.0000
tree
0.0032
0.0006
0.0454
0.9507
0.0000
0.0000
car
0.1395
0.0140
0.0090
0.0051
0.8325
0.0000
clutter
0.0117
0.9880
0.0004
0.0000
0.0000
0.0000
Precision/Correctness
0.916
0.794
0.821
0.936
0.892
0.000
Recall/Completeness
0.878
0.955
0.804
0.951
0.832
0.000
F1
0.897
0.867
0.812
0.943
0.861
-nan

Overall accuracy 0.882

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
0.9044
0.0240
0.0397
0.0273
0.0046
0.0000
building
0.0360
0.9466
0.0115
0.0058
0.0002
0.0000
low_veg
0.1111
0.0353
0.7243
0.1289
0.0004
0.0000
tree
0.0244
0.0035
0.0682
0.9038
0.0001
0.0000
car
0.3143
0.0100
0.0171
0.0047
0.6539
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.880
0.914
0.809
0.879
0.817
--
Recall/Completeness
0.904
0.947
0.724
0.904
0.654
--
F1
0.892
0.930
0.764
0.891
0.727
--

Overall accuracy 0.873

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
0.9341
0.0175
0.0321
0.0131
0.0031
0.0000
building
0.0295
0.9639
0.0039
0.0026
0.0001
0.0000
low_veg
0.1074
0.0235
0.7660
0.1027
0.0004
0.0000
tree
0.0102
0.0009
0.0516
0.9373
0.0000
0.0000
car
0.2152
0.0079
0.0119
0.0002
0.7648
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.907
0.943
0.852
0.916
0.847
--
Recall/Completeness
0.934
0.964
0.766
0.937
0.765
--
F1
0.920
0.953
0.807
0.926
0.804
--

Overall accuracy 0.906

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
0.8290
0.0346
0.0978
0.0351
0.0035
0.0000
building
0.0788
0.8610
0.0449
0.0146
0.0007
0.0000
low_veg
0.0343
0.0335
0.8504
0.0814
0.0004
0.0000
tree
0.0221
0.0069
0.1600
0.8110
0.0001
0.0000
car
0.2081
0.0036
0.0049
0.0034
0.7800
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.774
0.887
0.847
0.833
0.782
--
Recall/Completeness
0.829
0.861
0.850
0.811
0.780
--
F1
0.800
0.874
0.849
0.822
0.781
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8580
0.0245
0.0898
0.0255
0.0022
0.0000
building
0.0760
0.8723
0.0381
0.0129
0.0007
0.0000
low_veg
0.0307
0.0298
0.8678
0.0713
0.0004
0.0000
tree
0.0149
0.0034
0.1374
0.8443
0.0001
0.0000
car
0.1069
0.0021
0.0002
0.0003
0.8905
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.797
0.908
0.870
0.854
0.795
--
Recall/Completeness
0.858
0.872
0.868
0.844
0.890
--
F1
0.826
0.890
0.869
0.849
0.840
--

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9490
0.0237
0.0174
0.0057
0.0042
0.0000
building
0.1013
0.8670
0.0280
0.0032
0.0006
0.0000
low_veg
0.0898
0.0215
0.6950
0.1936
0.0001
0.0000
tree
0.0168
0.0034
0.0575
0.9221
0.0002
0.0000
car
0.3236
0.0165
0.0074
0.0041
0.6476
0.0007
clutter
0.7336
0.0607
0.1425
0.0030
0.0595
0.0007
Precision/Correctness
0.819
0.959
0.862
0.693
0.925
0.066
Recall/Completeness
0.949
0.867
0.695
0.922
0.648
0.001
F1
0.879
0.911
0.770
0.791
0.762
0.001

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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9707
0.0159
0.0107
0.0019
0.0008
0.0000
building
0.0901
0.8862
0.0219
0.0013
0.0004
0.0000
low_veg
0.0731
0.0110
0.7308
0.1851
0.0000
0.0000
tree
0.0067
0.0008
0.0385
0.9540
0.0000
0.0000
car
0.2195
0.0160
0.0047
0.0031
0.7560
0.0007
clutter
0.7732
0.0445
0.1227
0.0005
0.0581
0.0010
Precision/Correctness
0.851
0.977
0.899
0.720
0.972
0.090
Recall/Completeness
0.971
0.886
0.731
0.954
0.756
0.001
F1
0.907
0.929
0.806
0.821
0.850
0.002

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9483
0.0154
0.0225
0.0109
0.0028
0.0000
building
0.0469
0.9331
0.0147
0.0053
0.0000
0.0000
low_veg
0.0704
0.0212
0.6338
0.2744
0.0001
0.0000
tree
0.0079
0.0018
0.0301
0.9602
0.0000
0.0000
car
0.2801
0.0030
0.0053
0.0089
0.7025
0.0002
clutter
0.5050
0.4084
0.0207
0.0128
0.0416
0.0116
Precision/Correctness
0.904
0.920
0.843
0.690
0.793
0.972
Recall/Completeness
0.948
0.933
0.634
0.960
0.703
0.012
F1
0.926
0.927
0.724
0.803
0.745
0.023

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9675
0.0098
0.0162
0.0047
0.0018
0.0000
building
0.0413
0.9471
0.0091
0.0025
0.0000
0.0000
low_veg
0.0466
0.0109
0.6799
0.2625
0.0001
0.0000
tree
0.0026
0.0003
0.0221
0.9750
0.0000
0.0000
car
0.1837
0.0006
0.0041
0.0064
0.8051
0.0001
clutter
0.4843
0.4390
0.0146
0.0052
0.0438
0.0131
Precision/Correctness
0.926
0.938
0.883
0.728
0.806
0.994
Recall/Completeness
0.968
0.947
0.680
0.975
0.805
0.013
F1
0.946
0.943
0.768
0.833
0.806
0.026

Overall accuracy 0.895

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
0.9658
0.0168
0.0115
0.0025
0.0033
0.0000
building
0.1339
0.8498
0.0129
0.0014
0.0020
0.0001
low_veg
0.0428
0.0055
0.9233
0.0283
0.0001
0.0000
tree
0.0441
0.0030
0.1126
0.8397
0.0006
0.0000
car
0.1750
0.2484
0.0078
0.0036
0.5644
0.0008
clutter
0.5145
0.2977
0.0578
0.0064
0.0628
0.0609
Precision/Correctness
0.839
0.898
0.857
0.942
0.835
0.964
Recall/Completeness
0.966
0.850
0.923
0.840
0.564
0.061
F1
0.898
0.873
0.889
0.888
0.673
0.114

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9751
0.0143
0.0085
0.0004
0.0016
0.0000
building
0.1232
0.8691
0.0061
0.0001
0.0013
0.0001
low_veg
0.0226
0.0036
0.9497
0.0241
0.0000
0.0000
tree
0.0169
0.0008
0.0880
0.8942
0.0001
0.0000
car
0.0938
0.2784
0.0027
0.0009
0.6236
0.0006
clutter
0.4703
0.3351
0.0574
0.0017
0.0659
0.0695
Precision/Correctness
0.868
0.912
0.896
0.962
0.855
0.976
Recall/Completeness
0.975
0.869
0.950
0.894
0.624
0.070
F1
0.918
0.890
0.922
0.927
0.721
0.130

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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9695
0.0220
0.0058
0.0013
0.0013
0.0002
building
0.0881
0.8246
0.0846
0.0024
0.0002
0.0000
low_veg
0.0731
0.0194
0.7123
0.1951
0.0000
0.0000
tree
0.0532
0.0127
0.1578
0.7763
0.0000
0.0000
car
0.3172
0.0900
0.0056
0.0014
0.5858
0.0000
clutter
0.4939
0.2386
0.0023
0.0005
0.0672
0.1975
Precision/Correctness
0.836
0.899
0.716
0.722
0.642
0.991
Recall/Completeness
0.969
0.825
0.712
0.776
0.586
0.198
F1
0.898
0.860
0.714
0.748
0.612
0.329

Overall accuracy 0.821

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
0.9779
0.0183
0.0028
0.0002
0.0008
0.0000
building
0.0816
0.8335
0.0833
0.0013
0.0002
0.0000
low_veg
0.0403
0.0135
0.7367
0.2095
0.0000
0.0000
tree
0.0237
0.0090
0.1438
0.8235
0.0000
0.0000
car
0.2046
0.1077
0.0011
0.0003
0.6863
0.0000
clutter
0.4561
0.2525
0.0014
0.0004
0.0708
0.2189
Precision/Correctness
0.867
0.914
0.731
0.724
0.634
0.998
Recall/Completeness
0.978
0.834
0.737
0.824
0.686
0.219
F1
0.919
0.872
0.734
0.771
0.659
0.359

Overall accuracy 0.841

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
0.8991
0.0305
0.0587
0.0111
0.0005
0.0000
building
0.0929
0.8795
0.0230
0.0036
0.0010
0.0000
low_veg
0.0647
0.0093
0.7108
0.2149
0.0001
0.0000
tree
0.0378
0.0035
0.1380
0.8206
0.0001
0.0000
car
0.3316
0.0617
0.0029
0.0071
0.5967
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.812
0.954
0.768
0.771
0.907
--
Recall/Completeness
0.899
0.879
0.711
0.821
0.597
--
F1
0.853
0.915
0.738
0.795
0.720
--

Overall accuracy 0.827

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
0.9331
0.0263
0.0386
0.0020
0.0000
0.0000
building
0.0801
0.9066
0.0114
0.0009
0.0009
0.0000
low_veg
0.0483
0.0039
0.7347
0.2129
0.0001
0.0000
tree
0.0131
0.0008
0.1084
0.8777
0.0000
0.0000
car
0.2143
0.0646
0.0006
0.0021
0.7183
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.856
0.970
0.828
0.791
0.931
--
Recall/Completeness
0.933
0.907
0.735
0.878
0.718
--
F1
0.893
0.937
0.778
0.832
0.811
--

Overall accuracy 0.864

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9264
0.0209
0.0422
0.0082
0.0021
0.0001
building
0.0609
0.8807
0.0540
0.0036
0.0007
0.0000
low_veg
0.0443
0.0371
0.8331
0.0851
0.0003
0.0000
tree
0.0286
0.0027
0.1499
0.8188
0.0001
0.0000
car
0.2772
0.0543
0.0360
0.0063
0.6261
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.873
0.929
0.810
0.855
0.876
--
Recall/Completeness
0.926
0.881
0.833
0.819
0.626
--
F1
0.899
0.904
0.822
0.837
0.730
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9487
0.0180
0.0295
0.0025
0.0012
0.0001
building
0.0538
0.9036
0.0408
0.0013
0.0005
0.0000
low_veg
0.0269
0.0350
0.8636
0.0742
0.0003
0.0000
tree
0.0110
0.0009
0.1240
0.8642
0.0000
0.0000
car
0.1820
0.0640
0.0227
0.0007
0.7305
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.910
0.940
0.849
0.884
0.906
--
Recall/Completeness
0.949
0.904
0.864
0.864
0.730
--
F1
0.929
0.922
0.856
0.874
0.809
--

Overall accuracy 0.896

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image