Detailled semantic labeling (2D) result


Name G. Chen and X. Zhang
Affiliation Wuhan University, China
Abbreviation CVEO3
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
0.898
0.035
0.042
0.018
0.006
0.001
building
0.063
0.894
0.033
0.006
0.001
0.003
low_veg
0.055
0.018
0.763
0.164
0.001
0.000
tree
0.020
0.004
0.092
0.884
0.000
0.000
car
0.199
0.026
0.010
0.005
0.752
0.008
clutter
0.292
0.238
0.020
0.006
0.051
0.393
Precision/Correctness
0.869
0.932
0.799
0.827
0.787
0.749
Recall/Completeness
0.898
0.894
0.763
0.884
0.752
0.393
F1
0.883
0.913
0.781
0.854
0.769
0.515

Overall accuracy 0.858

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
0.925
0.028
0.034
0.008
0.004
0.001
building
0.057
0.910
0.027
0.003
0.001
0.003
low_veg
0.045
0.010
0.795
0.149
0.001
0.000
tree
0.009
0.002
0.076
0.913
0.000
0.000
car
0.104
0.028
0.006
0.002
0.850
0.010
clutter
0.257
0.254
0.017
0.002
0.050
0.420
Precision/Correctness
0.895
0.951
0.832
0.856
0.810
0.766
Recall/Completeness
0.925
0.910
0.795
0.913
0.850
0.420
F1
0.910
0.930
0.813
0.883
0.830
0.543

Overall accuracy 0.886
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9527
0.0082
0.0161
0.0144
0.0070
0.0016
building
0.0678
0.9001
0.0263
0.0037
0.0020
0.0001
low_veg
0.0749
0.0125
0.6937
0.2183
0.0006
0.0000
tree
0.0270
0.0030
0.0902
0.8775
0.0003
0.0021
car
0.2467
0.0010
0.0017
0.0011
0.7494
0.0000
clutter
0.0000
0.0000
0.0000
0.0060
0.0000
0.9940
Precision/Correctness
0.843
0.972
0.840
0.789
0.830
0.951
Recall/Completeness
0.953
0.900
0.694
0.877
0.749
0.994
F1
0.895
0.935
0.760
0.831
0.788
0.972

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
0.9752
0.0054
0.0084
0.0037
0.0056
0.0017
building
0.0606
0.9191
0.0174
0.0008
0.0019
0.0001
low_veg
0.0592
0.0058
0.7221
0.2123
0.0006
0.0000
tree
0.0125
0.0011
0.0775
0.9079
0.0001
0.0010
car
0.1177
0.0000
0.0003
0.0001
0.8819
0.0000
clutter
0.0000
0.0000
0.0000
0.0007
0.0000
0.9993
Precision/Correctness
0.879
0.986
0.871
0.813
0.842
0.966
Recall/Completeness
0.975
0.919
0.722
0.908
0.882
0.999
F1
0.925
0.952
0.790
0.858
0.861
0.982

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
0.9120
0.0611
0.0168
0.0040
0.0053
0.0008
building
0.0713
0.9061
0.0121
0.0021
0.0020
0.0064
low_veg
0.1263
0.0280
0.5258
0.3167
0.0032
0.0000
tree
0.0441
0.0146
0.0697
0.8714
0.0001
0.0000
car
0.2467
0.0103
0.0046
0.0020
0.7363
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.838
0.962
0.608
0.717
0.792
--
Recall/Completeness
0.912
0.906
0.526
0.871
0.736
--
F1
0.874
0.933
0.564
0.787
0.763
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9453
0.0361
0.0129
0.0013
0.0035
0.0009
building
0.0609
0.9187
0.0105
0.0012
0.0020
0.0066
low_veg
0.0899
0.0055
0.5522
0.3490
0.0034
0.0000
tree
0.0170
0.0060
0.0592
0.9177
0.0000
0.0000
car
0.1357
0.0043
0.0032
0.0006
0.8562
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.866
0.980
0.636
0.741
0.803
--
Recall/Completeness
0.945
0.919
0.552
0.918
0.856
--
F1
0.904
0.949
0.591
0.820
0.829
--

Overall accuracy 0.911

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area6_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9050
0.0532
0.0253
0.0078
0.0077
0.0010
building
0.0613
0.9001
0.0342
0.0028
0.0008
0.0008
low_veg
0.0965
0.0200
0.6994
0.1833
0.0008
0.0000
tree
0.0174
0.0039
0.0610
0.9176
0.0002
0.0000
car
0.2382
0.0304
0.0038
0.0010
0.7159
0.0106
clutter
--
--
--
--
--
--
Precision/Correctness
0.854
0.945
0.678
0.922
0.832
--
Recall/Completeness
0.905
0.900
0.699
0.918
0.716
--
F1
0.879
0.922
0.689
0.920
0.769
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9415
0.0333
0.0172
0.0018
0.0051
0.0011
building
0.0554
0.9121
0.0305
0.0007
0.0005
0.0007
low_veg
0.0808
0.0086
0.7427
0.1676
0.0004
0.0000
tree
0.0065
0.0014
0.0500
0.9421
0.0000
0.0000
car
0.1427
0.0220
0.0033
0.0000
0.8186
0.0134
clutter
--
--
--
--
--
--
Precision/Correctness
0.885
0.970
0.709
0.943
0.869
--
Recall/Completeness
0.942
0.912
0.743
0.942
0.819
--
F1
0.912
0.940
0.725
0.942
0.843
--

Overall accuracy 0.911

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8791
0.0314
0.0747
0.0075
0.0058
0.0015
building
0.0345
0.9361
0.0214
0.0036
0.0012
0.0032
low_veg
0.0393
0.0124
0.8703
0.0774
0.0006
0.0000
tree
0.0218
0.0054
0.0929
0.8790
0.0002
0.0007
car
0.2176
0.0024
0.0221
0.0024
0.7551
0.0004
clutter
--
--
--
--
--
--
Precision/Correctness
0.918
0.941
0.770
0.919
0.850
--
Recall/Completeness
0.879
0.936
0.870
0.879
0.755
--
F1
0.898
0.938
0.817
0.899
0.800
--

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
0.9061
0.0240
0.0633
0.0014
0.0038
0.0014
building
0.0266
0.9551
0.0127
0.0016
0.0009
0.0032
low_veg
0.0261
0.0053
0.8971
0.0710
0.0004
0.0000
tree
0.0080
0.0023
0.0710
0.9181
0.0000
0.0006
car
0.1115
0.0003
0.0147
0.0014
0.8718
0.0002
clutter
--
--
--
--
--
--
Precision/Correctness
0.950
0.962
0.813
0.938
0.882
--
Recall/Completeness
0.906
0.955
0.897
0.918
0.872
--
F1
0.928
0.958
0.853
0.928
0.877
--

Overall accuracy 0.919

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
0.8130
0.1110
0.0487
0.0216
0.0050
0.0007
building
0.0284
0.9377
0.0199
0.0126
0.0008
0.0006
low_veg
0.0799
0.0267
0.7010
0.1909
0.0015
0.0000
tree
0.0119
0.0074
0.0607
0.9197
0.0001
0.0000
car
0.1557
0.0282
0.0005
0.0017
0.8139
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.926
0.822
0.777
0.809
0.709
--
Recall/Completeness
0.813
0.938
0.701
0.920
0.814
--
F1
0.866
0.876
0.737
0.861
0.758
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8419
0.1030
0.0417
0.0092
0.0035
0.0007
building
0.0248
0.9487
0.0158
0.0097
0.0005
0.0004
low_veg
0.0777
0.0139
0.7449
0.1619
0.0016
0.0000
tree
0.0053
0.0036
0.0464
0.9447
0.0000
0.0000
car
0.0737
0.0327
0.0000
0.0000
0.8936
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.940
0.842
0.812
0.850
0.728
--
Recall/Completeness
0.842
0.949
0.745
0.945
0.894
--
F1
0.888
0.892
0.777
0.895
0.802
--

Overall accuracy 0.872

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
0.8647
0.0386
0.0586
0.0349
0.0030
0.0001
building
0.0989
0.8669
0.0196
0.0051
0.0000
0.0095
low_veg
0.0561
0.0163
0.7309
0.1961
0.0004
0.0002
tree
0.0094
0.0053
0.0895
0.8958
0.0000
0.0000
car
0.1626
0.0000
0.0032
0.0068
0.8274
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.757
0.932
0.824
0.841
0.880
--
Recall/Completeness
0.865
0.867
0.731
0.896
0.827
--
F1
0.807
0.898
0.775
0.867
0.853
--

Overall accuracy 0.838

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
0.8943
0.0363
0.0510
0.0178
0.0004
0.0002
building
0.1018
0.8747
0.0124
0.0009
0.0000
0.0102
low_veg
0.0540
0.0086
0.7654
0.1714
0.0004
0.0002
tree
0.0050
0.0018
0.0746
0.9187
0.0000
0.0000
car
0.0722
0.0000
0.0044
0.0025
0.9210
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.774
0.954
0.853
0.869
0.950
--
Recall/Completeness
0.894
0.875
0.765
0.919
0.921
--
F1
0.830
0.912
0.807
0.893
0.935
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7894
0.0720
0.0970
0.0360
0.0055
0.0002
building
0.0093
0.9641
0.0217
0.0049
0.0000
0.0000
low_veg
0.0780
0.0478
0.7338
0.1394
0.0008
0.0002
tree
0.0266
0.0097
0.1451
0.8186
0.0001
0.0000
car
0.1420
0.0000
0.0001
0.0041
0.8538
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.855
0.841
0.714
0.860
0.607
--
Recall/Completeness
0.789
0.964
0.734
0.819
0.854
--
F1
0.821
0.899
0.724
0.839
0.710
--

Overall accuracy 0.816

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
0.8072
0.0692
0.0933
0.0254
0.0048
0.0002
building
0.0062
0.9803
0.0121
0.0014
0.0000
0.0000
low_veg
0.0737
0.0338
0.7615
0.1301
0.0007
0.0002
tree
0.0187
0.0038
0.1316
0.8459
0.0000
0.0000
car
0.0357
0.0000
0.0000
0.0009
0.9634
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.879
0.869
0.735
0.881
0.593
--
Recall/Completeness
0.807
0.980
0.761
0.846
0.963
--
F1
0.842
0.921
0.748
0.863
0.734
--

Overall accuracy 0.840

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
0.8425
0.0226
0.0735
0.0580
0.0034
0.0000
building
0.0575
0.8673
0.0638
0.0114
0.0001
0.0000
low_veg
0.0354
0.0127
0.8069
0.1433
0.0001
0.0014
tree
0.0268
0.0032
0.1025
0.8676
0.0000
0.0000
car
0.0734
0.0055
0.0092
0.0053
0.9066
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.809
0.948
0.799
0.856
0.778
--
Recall/Completeness
0.842
0.867
0.807
0.868
0.907
--
F1
0.825
0.906
0.803
0.862
0.838
--

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_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8776
0.0179
0.0661
0.0367
0.0017
0.0000
building
0.0534
0.8887
0.0546
0.0033
0.0001
0.0000
low_veg
0.0317
0.0079
0.8315
0.1273
0.0001
0.0015
tree
0.0187
0.0009
0.0900
0.8904
0.0000
0.0000
car
0.0141
0.0010
0.0028
0.0008
0.9814
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.837
0.967
0.824
0.882
0.841
--
Recall/Completeness
0.878
0.889
0.831
0.890
0.981
--
F1
0.857
0.926
0.828
0.886
0.906
--

Overall accuracy 0.871

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8764
0.0252
0.0345
0.0563
0.0076
0.0000
building
0.0254
0.9344
0.0202
0.0172
0.0000
0.0028
low_veg
0.0585
0.0241
0.7076
0.2089
0.0009
0.0000
tree
0.0075
0.0017
0.0388
0.9519
0.0001
0.0000
car
0.1313
0.0000
0.0109
0.0171
0.8407
0.0000
clutter
0.0114
0.9883
0.0001
0.0003
0.0000
0.0000
Precision/Correctness
0.901
0.804
0.839
0.877
0.668
0.000
Recall/Completeness
0.876
0.934
0.708
0.952
0.841
0.000
F1
0.888
0.865
0.768
0.913
0.744
-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
0.9092
0.0165
0.0326
0.0360
0.0058
0.0000
building
0.0230
0.9463
0.0157
0.0121
0.0000
0.0030
low_veg
0.0534
0.0133
0.7532
0.1791
0.0010
0.0000
tree
0.0032
0.0005
0.0293
0.9671
0.0000
0.0000
car
0.0669
0.0000
0.0019
0.0095
0.9217
0.0000
clutter
0.0089
0.9911
0.0000
0.0000
0.0000
0.0000
Precision/Correctness
0.922
0.823
0.867
0.905
0.668
0.000
Recall/Completeness
0.909
0.946
0.753
0.967
0.922
0.000
F1
0.916
0.880
0.806
0.935
0.774
-nan

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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8953
0.0228
0.0430
0.0307
0.0081
0.0000
building
0.0424
0.9379
0.0120
0.0069
0.0007
0.0000
low_veg
0.1130
0.0302
0.6885
0.1671
0.0012
0.0000
tree
0.0208
0.0025
0.0548
0.9216
0.0002
0.0000
car
0.2193
0.0006
0.0122
0.0086
0.7559
0.0035
clutter
--
--
--
--
--
--
Precision/Correctness
0.880
0.922
0.814
0.856
0.731
--
Recall/Completeness
0.895
0.938
0.688
0.922
0.756
--
F1
0.888
0.930
0.746
0.887
0.743
--

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
0.9269
0.0164
0.0368
0.0138
0.0061
0.0000
building
0.0372
0.9541
0.0051
0.0032
0.0005
0.0000
low_veg
0.1110
0.0193
0.7306
0.1378
0.0013
0.0000
tree
0.0084
0.0003
0.0399
0.9513
0.0000
0.0000
car
0.1241
0.0000
0.0071
0.0015
0.8630
0.0043
clutter
--
--
--
--
--
--
Precision/Correctness
0.904
0.950
0.854
0.895
0.751
--
Recall/Completeness
0.927
0.954
0.731
0.951
0.863
--
F1
0.915
0.952
0.788
0.922
0.803
--

Overall accuracy 0.900

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
0.8247
0.0379
0.0903
0.0417
0.0054
0.0000
building
0.0518
0.8840
0.0453
0.0181
0.0006
0.0002
low_veg
0.0356
0.0176
0.8487
0.0977
0.0004
0.0000
tree
0.0192
0.0061
0.1461
0.8281
0.0005
0.0000
car
0.1728
0.0000
0.0020
0.0016
0.8235
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.801
0.921
0.856
0.809
0.726
--
Recall/Completeness
0.825
0.884
0.849
0.828
0.824
--
F1
0.812
0.902
0.852
0.818
0.772
--

Overall accuracy 0.847

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8557
0.0267
0.0838
0.0301
0.0038
0.0000
building
0.0502
0.8953
0.0389
0.0148
0.0006
0.0002
low_veg
0.0327
0.0130
0.8673
0.0866
0.0004
0.0000
tree
0.0123
0.0030
0.1261
0.8582
0.0005
0.0000
car
0.0784
0.0000
0.0000
0.0000
0.9216
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.822
0.945
0.877
0.832
0.729
--
Recall/Completeness
0.856
0.895
0.867
0.858
0.922
--
F1
0.838
0.919
0.872
0.845
0.814
--

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
0.9322
0.0249
0.0240
0.0080
0.0092
0.0017
building
0.0927
0.8737
0.0283
0.0037
0.0015
0.0002
low_veg
0.0669
0.0195
0.6820
0.2298
0.0014
0.0005
tree
0.0129
0.0038
0.0631
0.9200
0.0002
0.0000
car
0.2202
0.0052
0.0080
0.0051
0.7564
0.0050
clutter
0.4966
0.0942
0.1212
0.0082
0.1625
0.1173
Precision/Correctness
0.844
0.960
0.845
0.653
0.855
0.256
Recall/Completeness
0.932
0.874
0.682
0.920
0.756
0.117
F1
0.886
0.915
0.755
0.764
0.803
0.161

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9594
0.0151
0.0167
0.0028
0.0042
0.0017
building
0.0835
0.8915
0.0225
0.0010
0.0012
0.0002
low_veg
0.0515
0.0095
0.7178
0.2195
0.0012
0.0005
tree
0.0048
0.0007
0.0457
0.9488
0.0000
0.0000
car
0.1207
0.0032
0.0058
0.0036
0.8604
0.0063
clutter
0.4829
0.0836
0.0892
0.0053
0.1846
0.1544
Precision/Correctness
0.873
0.979
0.881
0.683
0.899
0.258
Recall/Completeness
0.959
0.892
0.718
0.949
0.860
0.154
F1
0.914
0.933
0.791
0.795
0.879
0.193

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9284
0.0169
0.0332
0.0145
0.0063
0.0007
building
0.0342
0.9355
0.0182
0.0069
0.0002
0.0051
low_veg
0.0436
0.0145
0.6046
0.3368
0.0005
0.0000
tree
0.0040
0.0011
0.0278
0.9670
0.0000
0.0000
car
0.2172
0.0055
0.0009
0.0107
0.7367
0.0290
clutter
0.3780
0.3257
0.0320
0.0111
0.0823
0.1709
Precision/Correctness
0.929
0.930
0.796
0.645
0.653
0.709
Recall/Completeness
0.928
0.935
0.605
0.967
0.737
0.171
F1
0.928
0.933
0.687
0.774
0.692
0.275

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
0.9522
0.0111
0.0253
0.0060
0.0049
0.0005
building
0.0298
0.9500
0.0116
0.0032
0.0001
0.0053
low_veg
0.0271
0.0049
0.6417
0.3259
0.0003
0.0000
tree
0.0009
0.0001
0.0208
0.9782
0.0000
0.0000
car
0.1241
0.0049
0.0000
0.0073
0.8282
0.0354
clutter
0.3572
0.3380
0.0259
0.0040
0.0823
0.1927
Precision/Correctness
0.947
0.948
0.837
0.683
0.657
0.715
Recall/Completeness
0.952
0.950
0.642
0.978
0.828
0.193
F1
0.950
0.949
0.727
0.804
0.733
0.304

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9616
0.0098
0.0153
0.0041
0.0075
0.0017
building
0.1374
0.8329
0.0180
0.0028
0.0035
0.0054
low_veg
0.0275
0.0010
0.9232
0.0480
0.0003
0.0000
tree
0.0314
0.0019
0.0714
0.8948
0.0004
0.0001
car
0.1272
0.0936
0.0036
0.0057
0.7446
0.0253
clutter
0.5530
0.0974
0.0524
0.0109
0.0784
0.2079
Precision/Correctness
0.845
0.955
0.868
0.910
0.802
0.695
Recall/Completeness
0.962
0.833
0.923
0.895
0.745
0.208
F1
0.899
0.890
0.895
0.902
0.772
0.320

Overall accuracy 0.879

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9754
0.0077
0.0114
0.0007
0.0039
0.0010
building
0.1299
0.8494
0.0125
0.0005
0.0030
0.0048
low_veg
0.0135
0.0001
0.9465
0.0398
0.0001
0.0000
tree
0.0099
0.0002
0.0556
0.9343
0.0000
0.0000
car
0.0550
0.1020
0.0001
0.0012
0.8106
0.0311
clutter
0.5220
0.1097
0.0540
0.0043
0.0797
0.2303
Precision/Correctness
0.867
0.964
0.900
0.940
0.829
0.723
Recall/Completeness
0.975
0.849
0.946
0.934
0.811
0.230
F1
0.918
0.903
0.923
0.937
0.820
0.349

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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9723
0.0089
0.0081
0.0033
0.0021
0.0054
building
0.0694
0.8424
0.0706
0.0093
0.0027
0.0056
low_veg
0.0545
0.0126
0.8010
0.1298
0.0011
0.0011
tree
0.0387
0.0021
0.1145
0.8445
0.0000
0.0001
car
0.2301
0.0567
0.0049
0.0021
0.6978
0.0084
clutter
0.3324
0.0351
0.0003
0.0021
0.0568
0.5734
Precision/Correctness
0.875
0.968
0.776
0.786
0.650
0.886
Recall/Completeness
0.972
0.842
0.801
0.845
0.698
0.573
F1
0.921
0.901
0.788
0.814
0.673
0.696

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_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9867
0.0054
0.0038
0.0007
0.0010
0.0024
building
0.0628
0.8528
0.0699
0.0074
0.0024
0.0047
low_veg
0.0258
0.0068
0.8325
0.1327
0.0010
0.0011
tree
0.0148
0.0001
0.1038
0.8812
0.0000
0.0000
car
0.1248
0.0696
0.0020
0.0006
0.7939
0.0092
clutter
0.2939
0.0357
0.0001
0.0005
0.0596
0.6102
Precision/Correctness
0.904
0.979
0.792
0.800
0.651
0.922
Recall/Completeness
0.987
0.853
0.833
0.881
0.794
0.610
F1
0.944
0.911
0.812
0.838
0.715
0.734

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8791
0.0340
0.0703
0.0142
0.0014
0.0009
building
0.0679
0.9077
0.0186
0.0049
0.0008
0.0001
low_veg
0.0419
0.0115
0.7272
0.2187
0.0007
0.0000
tree
0.0263
0.0045
0.1432
0.8257
0.0003
0.0000
car
0.2248
0.0168
0.0013
0.0036
0.7534
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.859
0.951
0.763
0.766
0.874
--
Recall/Completeness
0.879
0.908
0.727
0.826
0.753
--
F1
0.869
0.929
0.744
0.795
0.809
--

Overall accuracy 0.836

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
0.9187
0.0283
0.0487
0.0029
0.0003
0.0011
building
0.0572
0.9333
0.0079
0.0008
0.0006
0.0001
low_veg
0.0294
0.0051
0.7483
0.2167
0.0006
0.0000
tree
0.0070
0.0010
0.1146
0.8772
0.0001
0.0000
car
0.1270
0.0219
0.0001
0.0000
0.8510
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.898
0.968
0.819
0.788
0.920
--
Recall/Completeness
0.919
0.933
0.748
0.877
0.851
--
F1
0.908
0.951
0.782
0.830
0.884
--

Overall accuracy 0.872

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
0.8934
0.0285
0.0587
0.0129
0.0060
0.0004
building
0.0479
0.8533
0.0907
0.0058
0.0013
0.0010
low_veg
0.0276
0.0114
0.8506
0.1099
0.0004
0.0000
tree
0.0177
0.0022
0.1177
0.8623
0.0001
0.0000
car
0.1719
0.0384
0.0507
0.0050
0.7339
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.906
0.949
0.793
0.825
0.771
--
Recall/Completeness
0.893
0.853
0.851
0.862
0.734
--
F1
0.899
0.899
0.821
0.843
0.752
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9234
0.0266
0.0415
0.0036
0.0045
0.0005
building
0.0423
0.8773
0.0763
0.0020
0.0010
0.0010
low_veg
0.0164
0.0079
0.8817
0.0936
0.0004
0.0000
tree
0.0054
0.0008
0.0961
0.8977
0.0000
0.0000
car
0.0877
0.0467
0.0352
0.0009
0.8295
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.934
0.959
0.830
0.862
0.793
--
Recall/Completeness
0.923
0.877
0.882
0.898
0.830
--
F1
0.929
0.917
0.855
0.879
0.811
--

Overall accuracy 0.894

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image