Detailled semantic labeling (2D) result


Name T. Quang et al.
Affiliation HUST, Vietnam
Abbreviation HUST
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
0.907
0.043
0.032
0.017
0.001
0.000
building
0.080
0.899
0.008
0.012
0.001
0.000
low_veg
0.112
0.038
0.688
0.162
0.000
0.000
tree
0.032
0.012
0.096
0.860
0.000
0.000
car
0.743
0.092
0.012
0.006
0.147
0.000
clutter
0.811
0.162
0.017
0.005
0.005
0.000
Precision/Correctness
0.785
0.902
0.818
0.819
0.792
-nan
Recall/Completeness
0.907
0.899
0.688
0.860
0.147
0.000
F1
0.842
0.900
0.747
0.839
0.247
-nan

Overall accuracy 0.829

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
0.036
0.023
0.010
0.000
0.000
building
0.070
0.915
0.005
0.008
0.001
0.000
low_veg
0.101
0.029
0.722
0.148
0.000
0.000
tree
0.021
0.008
0.080
0.891
0.000
0.000
car
0.699
0.110
0.010
0.003
0.178
0.000
clutter
0.811
0.169
0.014
0.001
0.005
0.000
Precision/Correctness
0.814
0.924
0.855
0.847
0.795
-nan
Recall/Completeness
0.931
0.915
0.722
0.891
0.178
0.000
F1
0.869
0.920
0.783
0.869
0.290
-nan

Overall accuracy 0.859
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9467
0.0164
0.0263
0.0099
0.0007
0.0000
building
0.0445
0.9401
0.0106
0.0046
0.0001
0.0000
low_veg
0.2278
0.0320
0.6138
0.1264
0.0001
0.0000
tree
0.0509
0.0173
0.0931
0.8386
0.0000
0.0000
car
0.8440
0.0250
0.0133
0.0020
0.1157
0.0000
clutter
0.9743
0.0140
0.0009
0.0108
0.0000
0.0000
Precision/Correctness
0.683
0.928
0.823
0.856
0.894
-nan
Recall/Completeness
0.947
0.940
0.614
0.839
0.116
0.000
F1
0.794
0.934
0.703
0.847
0.205
-nan

Overall accuracy 0.807

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
0.0117
0.0126
0.0032
0.0006
0.0000
building
0.0327
0.9580
0.0074
0.0019
0.0001
0.0000
low_veg
0.2202
0.0224
0.6365
0.1208
0.0001
0.0000
tree
0.0352
0.0142
0.0787
0.8719
0.0000
0.0000
car
0.8104
0.0317
0.0137
0.0006
0.1436
0.0000
clutter
0.9856
0.0120
0.0003
0.0021
0.0000
0.0000
Precision/Correctness
0.709
0.948
0.861
0.878
0.897
-nan
Recall/Completeness
0.972
0.958
0.637
0.872
0.144
0.000
F1
0.820
0.953
0.732
0.875
0.248
-nan

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8753
0.1159
0.0023
0.0061
0.0004
0.0000
building
0.0607
0.9338
0.0014
0.0041
0.0000
0.0000
low_veg
0.3759
0.1023
0.2574
0.2644
0.0000
0.0000
tree
0.0846
0.0381
0.0361
0.8412
0.0000
0.0000
car
0.8201
0.0704
0.0044
0.0022
0.1030
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.793
0.927
0.778
0.717
0.916
--
Recall/Completeness
0.875
0.934
0.257
0.841
0.103
--
F1
0.832
0.930
0.387
0.774
0.185
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8963
0.0987
0.0010
0.0037
0.0004
0.0000
building
0.0454
0.9504
0.0013
0.0029
0.0000
0.0000
low_veg
0.3620
0.0735
0.2853
0.2792
0.0000
0.0000
tree
0.0560
0.0233
0.0321
0.8886
0.0000
0.0000
car
0.7935
0.0736
0.0051
0.0009
0.1269
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.829
0.944
0.815
0.750
0.930
--
Recall/Completeness
0.896
0.950
0.285
0.889
0.127
--
F1
0.862
0.947
0.423
0.813
0.223
--

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_area6_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8846
0.0831
0.0211
0.0107
0.0005
0.0000
building
0.0479
0.9399
0.0059
0.0058
0.0004
0.0000
low_veg
0.0871
0.1022
0.5395
0.2710
0.0001
0.0000
tree
0.0192
0.0110
0.0434
0.9263
0.0001
0.0000
car
0.7931
0.0613
0.0073
0.0032
0.1351
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.838
0.896
0.743
0.888
0.891
--
Recall/Completeness
0.885
0.940
0.540
0.926
0.135
--
F1
0.861
0.917
0.625
0.907
0.235
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9153
0.0663
0.0135
0.0047
0.0003
0.0000
building
0.0360
0.9570
0.0042
0.0025
0.0003
0.0000
low_veg
0.0744
0.0874
0.5812
0.2569
0.0000
0.0000
tree
0.0097
0.0061
0.0336
0.9505
0.0000
0.0000
car
0.7627
0.0595
0.0091
0.0007
0.1680
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.877
0.923
0.791
0.913
0.928
--
Recall/Completeness
0.915
0.957
0.581
0.951
0.168
--
F1
0.896
0.939
0.670
0.931
0.285
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9014
0.0554
0.0304
0.0123
0.0005
0.0000
building
0.0629
0.9209
0.0094
0.0067
0.0001
0.0000
low_veg
0.1355
0.0243
0.7665
0.0737
0.0001
0.0000
tree
0.0453
0.0141
0.0841
0.8565
0.0000
0.0000
car
0.7551
0.0460
0.0104
0.0060
0.1824
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.814
0.890
0.825
0.911
0.944
--
Recall/Completeness
0.901
0.921
0.766
0.857
0.182
--
F1
0.855
0.905
0.795
0.883
0.306
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9245
0.0491
0.0206
0.0055
0.0003
0.0000
building
0.0474
0.9425
0.0059
0.0041
0.0001
0.0000
low_veg
0.1189
0.0156
0.7994
0.0661
0.0001
0.0000
tree
0.0298
0.0086
0.0667
0.8948
0.0000
0.0000
car
0.7072
0.0546
0.0061
0.0050
0.2271
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.851
0.914
0.866
0.932
0.956
--
Recall/Completeness
0.924
0.943
0.799
0.895
0.227
--
F1
0.886
0.928
0.831
0.913
0.367
--

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_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8716
0.0778
0.0302
0.0192
0.0012
0.0000
building
0.0219
0.9565
0.0075
0.0127
0.0014
0.0000
low_veg
0.1172
0.0644
0.6585
0.1599
0.0001
0.0000
tree
0.0246
0.0115
0.0980
0.8659
0.0000
0.0000
car
0.7512
0.0765
0.0063
0.0071
0.1589
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.900
0.839
0.777
0.823
0.590
--
Recall/Completeness
0.872
0.956
0.658
0.866
0.159
--
F1
0.886
0.894
0.713
0.844
0.250
--

Overall accuracy 0.848

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
0.9012
0.0660
0.0223
0.0094
0.0011
0.0000
building
0.0143
0.9708
0.0049
0.0088
0.0013
0.0000
low_veg
0.1146
0.0527
0.6983
0.1343
0.0001
0.0000
tree
0.0158
0.0064
0.0776
0.9002
0.0000
0.0000
car
0.7138
0.0888
0.0050
0.0025
0.1899
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.921
0.867
0.823
0.863
0.575
--
Recall/Completeness
0.901
0.971
0.698
0.900
0.190
--
F1
0.911
0.916
0.756
0.881
0.285
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8567
0.0353
0.0652
0.0422
0.0007
0.0000
building
0.1791
0.8037
0.0068
0.0102
0.0002
0.0000
low_veg
0.0656
0.0488
0.6320
0.2522
0.0014
0.0000
tree
0.0097
0.0106
0.1098
0.8700
0.0000
0.0000
car
0.6726
0.0766
0.0223
0.0118
0.2167
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.669
0.867
0.781
0.799
0.658
--
Recall/Completeness
0.857
0.804
0.632
0.870
0.217
--
F1
0.751
0.834
0.699
0.833
0.326
--

Overall accuracy 0.783

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
0.8885
0.0322
0.0519
0.0269
0.0005
0.0000
building
0.1866
0.8053
0.0039
0.0039
0.0003
0.0000
low_veg
0.0646
0.0403
0.6614
0.2321
0.0015
0.0000
tree
0.0051
0.0067
0.0947
0.8935
0.0000
0.0000
car
0.6089
0.1034
0.0183
0.0142
0.2553
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.684
0.892
0.812
0.825
0.629
--
Recall/Completeness
0.889
0.805
0.661
0.893
0.255
--
F1
0.773
0.847
0.729
0.858
0.363
--

Overall accuracy 0.807

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
0.8787
0.0319
0.0562
0.0330
0.0002
0.0000
building
0.0308
0.9536
0.0087
0.0063
0.0006
0.0000
low_veg
0.1287
0.0710
0.6374
0.1625
0.0003
0.0000
tree
0.0467
0.0227
0.1124
0.8181
0.0000
0.0000
car
0.7081
0.0293
0.0055
0.0112
0.2458
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.784
0.835
0.757
0.846
0.731
--
Recall/Completeness
0.879
0.954
0.637
0.818
0.246
--
F1
0.828
0.890
0.692
0.832
0.368
--

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_area14_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8989
0.0279
0.0495
0.0236
0.0001
0.0000
building
0.0219
0.9707
0.0047
0.0022
0.0005
0.0000
low_veg
0.1228
0.0570
0.6686
0.1514
0.0003
0.0000
tree
0.0386
0.0154
0.1023
0.8438
0.0000
0.0000
car
0.6540
0.0364
0.0036
0.0078
0.2983
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.811
0.867
0.780
0.868
0.764
--
Recall/Completeness
0.899
0.971
0.669
0.844
0.298
--
F1
0.853
0.916
0.720
0.856
0.429
--

Overall accuracy 0.834

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
0.8148
0.0363
0.0840
0.0645
0.0004
0.0000
building
0.0373
0.9317
0.0120
0.0189
0.0000
0.0000
low_veg
0.0353
0.0265
0.7322
0.2053
0.0007
0.0000
tree
0.0231
0.0100
0.1145
0.8523
0.0001
0.0000
car
0.5683
0.0473
0.0712
0.0125
0.3007
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.824
0.906
0.791
0.808
0.686
--
Recall/Completeness
0.815
0.932
0.732
0.852
0.301
--
F1
0.820
0.918
0.761
0.830
0.418
--

Overall accuracy 0.824

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8504
0.0316
0.0734
0.0442
0.0003
0.0000
building
0.0324
0.9525
0.0055
0.0097
0.0000
0.0000
low_veg
0.0312
0.0199
0.7561
0.1920
0.0008
0.0000
tree
0.0164
0.0064
0.1023
0.8748
0.0001
0.0000
car
0.4742
0.0654
0.0587
0.0094
0.3923
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.854
0.929
0.819
0.831
0.668
--
Recall/Completeness
0.850
0.952
0.756
0.875
0.392
--
F1
0.852
0.940
0.786
0.852
0.494
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8280
0.0371
0.0990
0.0357
0.0002
0.0000
building
0.0235
0.9403
0.0242
0.0120
0.0000
0.0000
low_veg
0.0751
0.0275
0.6833
0.2136
0.0006
0.0000
tree
0.0082
0.0026
0.0515
0.9377
0.0000
0.0000
car
0.6002
0.0891
0.0326
0.0166
0.2615
0.0000
clutter
0.9481
0.0072
0.0436
0.0011
0.0000
0.0000
Precision/Correctness
0.770
0.914
0.741
0.884
0.879
-nan
Recall/Completeness
0.828
0.940
0.683
0.938
0.261
0.000
F1
0.798
0.927
0.711
0.910
0.403
-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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8586
0.0308
0.0900
0.0207
0.0000
0.0000
building
0.0181
0.9556
0.0212
0.0052
0.0000
0.0000
low_veg
0.0708
0.0195
0.7240
0.1851
0.0006
0.0000
tree
0.0039
0.0011
0.0387
0.9563
0.0000
0.0000
car
0.5461
0.0985
0.0255
0.0086
0.3212
0.0000
clutter
0.9589
0.0073
0.0332
0.0006
0.0000
0.0000
Precision/Correctness
0.788
0.937
0.775
0.910
0.887
-nan
Recall/Completeness
0.859
0.956
0.724
0.956
0.321
0.000
F1
0.822
0.946
0.749
0.933
0.472
-nan

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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8890
0.0378
0.0371
0.0354
0.0007
0.0000
building
0.0385
0.9417
0.0095
0.0103
0.0000
0.0000
low_veg
0.1444
0.0499
0.6116
0.1937
0.0002
0.0000
tree
0.0300
0.0060
0.0701
0.8939
0.0000
0.0000
car
0.7836
0.0603
0.0156
0.0123
0.1281
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.842
0.872
0.784
0.831
0.845
--
Recall/Completeness
0.889
0.942
0.612
0.894
0.128
--
F1
0.865
0.906
0.687
0.861
0.223
--

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_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9171
0.0320
0.0298
0.0206
0.0005
0.0000
building
0.0284
0.9612
0.0058
0.0046
0.0000
0.0000
low_veg
0.1408
0.0391
0.6488
0.1709
0.0003
0.0000
tree
0.0169
0.0031
0.0553
0.9247
0.0000
0.0000
car
0.7457
0.0771
0.0122
0.0071
0.1579
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.872
0.899
0.826
0.866
0.858
--
Recall/Completeness
0.917
0.961
0.649
0.925
0.158
--
F1
0.894
0.929
0.727
0.895
0.267
--

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_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8421
0.0442
0.0779
0.0350
0.0007
0.0000
building
0.0748
0.8852
0.0146
0.0254
0.0000
0.0000
low_veg
0.0484
0.0247
0.8456
0.0811
0.0002
0.0000
tree
0.0266
0.0126
0.1907
0.7700
0.0001
0.0000
car
0.7006
0.0003
0.0044
0.0063
0.2885
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.739
0.894
0.848
0.819
0.837
--
Recall/Completeness
0.842
0.885
0.846
0.770
0.288
--
F1
0.787
0.889
0.847
0.794
0.429
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8704
0.0358
0.0690
0.0245
0.0004
0.0000
building
0.0693
0.8985
0.0110
0.0213
0.0000
0.0000
low_veg
0.0440
0.0202
0.8655
0.0701
0.0002
0.0000
tree
0.0203
0.0092
0.1717
0.7987
0.0001
0.0000
car
0.6421
0.0001
0.0008
0.0043
0.3527
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.764
0.915
0.868
0.844
0.853
--
Recall/Completeness
0.870
0.898
0.865
0.799
0.353
--
F1
0.814
0.907
0.867
0.821
0.499
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9409
0.0356
0.0167
0.0058
0.0011
0.0000
building
0.0771
0.9065
0.0044
0.0115
0.0005
0.0000
low_veg
0.2132
0.0501
0.5667
0.1693
0.0008
0.0000
tree
0.0357
0.0162
0.0517
0.8965
0.0000
0.0000
car
0.7821
0.0372
0.0080
0.0055
0.1673
0.0000
clutter
0.8268
0.1189
0.0491
0.0051
0.0001
0.0000
Precision/Correctness
0.751
0.929
0.899
0.696
0.892
-nan
Recall/Completeness
0.941
0.907
0.567
0.896
0.167
0.000
F1
0.835
0.918
0.695
0.784
0.282
-nan

Overall accuracy 0.824

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9612
0.0259
0.0100
0.0025
0.0004
0.0000
building
0.0610
0.9274
0.0022
0.0090
0.0004
0.0000
low_veg
0.2009
0.0377
0.6021
0.1583
0.0009
0.0000
tree
0.0238
0.0105
0.0361
0.9296
0.0000
0.0000
car
0.7483
0.0385
0.0070
0.0044
0.2018
0.0000
clutter
0.8497
0.1144
0.0346
0.0012
0.0000
0.0000
Precision/Correctness
0.787
0.951
0.936
0.727
0.920
-nan
Recall/Completeness
0.961
0.927
0.602
0.930
0.202
0.000
F1
0.866
0.939
0.733
0.816
0.331
-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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9554
0.0198
0.0139
0.0106
0.0002
0.0000
building
0.0765
0.9080
0.0024
0.0091
0.0040
0.0000
low_veg
0.1219
0.0446
0.5503
0.2832
0.0000
0.0000
tree
0.0195
0.0070
0.0296
0.9439
0.0000
0.0000
car
0.8099
0.0333
0.0172
0.0104
0.1292
0.0000
clutter
0.7412
0.2262
0.0145
0.0088
0.0094
0.0000
Precision/Correctness
0.846
0.915
0.886
0.677
0.543
-nan
Recall/Completeness
0.955
0.908
0.550
0.944
0.129
0.000
F1
0.897
0.911
0.679
0.789
0.209
-nan

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9734
0.0142
0.0074
0.0050
0.0001
0.0000
building
0.0674
0.9222
0.0010
0.0052
0.0042
0.0000
low_veg
0.1037
0.0307
0.5916
0.2740
0.0000
0.0000
tree
0.0122
0.0040
0.0218
0.9620
0.0000
0.0000
car
0.7783
0.0371
0.0192
0.0078
0.1576
0.0000
clutter
0.7359
0.2411
0.0111
0.0018
0.0101
0.0000
Precision/Correctness
0.871
0.936
0.928
0.713
0.517
-nan
Recall/Completeness
0.973
0.922
0.592
0.962
0.158
0.000
F1
0.919
0.929
0.723
0.819
0.241
-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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9749
0.0123
0.0094
0.0026
0.0007
0.0000
building
0.2888
0.7004
0.0035
0.0039
0.0034
0.0000
low_veg
0.0751
0.0063
0.8787
0.0397
0.0003
0.0000
tree
0.0670
0.0058
0.0696
0.8576
0.0000
0.0000
car
0.5451
0.3000
0.0089
0.0041
0.1419
0.0000
clutter
0.5941
0.3707
0.0287
0.0038
0.0026
0.0000
Precision/Correctness
0.733
0.867
0.905
0.922
0.791
-nan
Recall/Completeness
0.975
0.700
0.879
0.858
0.142
0.000
F1
0.837
0.775
0.891
0.889
0.241
-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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9833
0.0096
0.0060
0.0006
0.0005
0.0000
building
0.2801
0.7139
0.0017
0.0010
0.0033
0.0000
low_veg
0.0526
0.0040
0.9102
0.0330
0.0003
0.0000
tree
0.0398
0.0033
0.0491
0.9079
0.0000
0.0000
car
0.4741
0.3529
0.0059
0.0013
0.1658
0.0000
clutter
0.5495
0.4193
0.0279
0.0008
0.0025
0.0000
Precision/Correctness
0.756
0.883
0.937
0.947
0.790
-nan
Recall/Completeness
0.983
0.714
0.910
0.908
0.166
0.000
F1
0.855
0.789
0.923
0.927
0.274
-nan

Overall accuracy 0.832

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
0.9784
0.0113
0.0079
0.0021
0.0002
0.0000
building
0.2258
0.7031
0.0121
0.0584
0.0007
0.0000
low_veg
0.1066
0.0219
0.7994
0.0721
0.0001
0.0000
tree
0.0570
0.0194
0.1654
0.7582
0.0001
0.0000
car
0.8403
0.0257
0.0098
0.0012
0.1229
0.0000
clutter
0.8601
0.1262
0.0022
0.0009
0.0106
0.0000
Precision/Correctness
0.722
0.923
0.836
0.733
0.647
-nan
Recall/Completeness
0.978
0.703
0.799
0.758
0.123
0.000
F1
0.831
0.798
0.817
0.746
0.207
-nan

Overall accuracy 0.784

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
0.9864
0.0090
0.0036
0.0008
0.0001
0.0000
building
0.2203
0.7108
0.0102
0.0582
0.0006
0.0000
low_veg
0.0723
0.0173
0.8361
0.0743
0.0000
0.0000
tree
0.0324
0.0154
0.1524
0.7997
0.0000
0.0000
car
0.8039
0.0306
0.0109
0.0004
0.1542
0.0000
clutter
0.8518
0.1351
0.0015
0.0003
0.0113
0.0000
Precision/Correctness
0.743
0.936
0.862
0.738
0.651
-nan
Recall/Completeness
0.986
0.711
0.836
0.800
0.154
0.000
F1
0.848
0.808
0.849
0.768
0.249
-nan

Overall accuracy 0.804

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
0.8863
0.0525
0.0459
0.0150
0.0003
0.0000
building
0.0667
0.9185
0.0083
0.0063
0.0001
0.0000
low_veg
0.1449
0.0206
0.6050
0.2290
0.0004
0.0000
tree
0.0512
0.0053
0.1194
0.8241
0.0000
0.0000
car
0.7495
0.1343
0.0057
0.0081
0.1024
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.757
0.923
0.783
0.756
0.774
--
Recall/Completeness
0.886
0.919
0.605
0.824
0.102
--
F1
0.817
0.921
0.682
0.789
0.181
--

Overall accuracy 0.805

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
0.9243
0.0478
0.0225
0.0050
0.0003
0.0000
building
0.0488
0.9464
0.0028
0.0019
0.0001
0.0000
low_veg
0.1286
0.0137
0.6328
0.2243
0.0005
0.0000
tree
0.0317
0.0015
0.0890
0.8778
0.0000
0.0000
car
0.6977
0.1648
0.0057
0.0040
0.1278
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.801
0.942
0.853
0.779
0.757
--
Recall/Completeness
0.924
0.946
0.633
0.878
0.128
--
F1
0.858
0.944
0.727
0.826
0.219
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9010
0.0406
0.0383
0.0194
0.0007
0.0000
building
0.0329
0.9340
0.0143
0.0176
0.0012
0.0000
low_veg
0.0664
0.0263
0.7717
0.1356
0.0001
0.0000
tree
0.0336
0.0117
0.1155
0.8391
0.0001
0.0000
car
0.7718
0.0889
0.0262
0.0070
0.1060
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.855
0.917
0.855
0.775
0.666
--
Recall/Completeness
0.901
0.934
0.772
0.839
0.106
--
F1
0.878
0.925
0.811
0.806
0.183
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9308
0.0373
0.0217
0.0096
0.0005
0.0000
building
0.0217
0.9579
0.0091
0.0102
0.0011
0.0000
low_veg
0.0547
0.0188
0.8097
0.1167
0.0001
0.0000
tree
0.0247
0.0084
0.0888
0.8781
0.0001
0.0000
car
0.7434
0.1086
0.0187
0.0019
0.1273
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.890
0.935
0.899
0.816
0.662
--
Recall/Completeness
0.931
0.958
0.810
0.878
0.127
--
F1
0.910
0.946
0.852
0.846
0.214
--

Overall accuracy 0.889

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image