Detailled semantic labeling (2D) result


Name W. Tang
Affiliation Wuhan University, China
Abbreviation WUH_W4
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
0.908
0.031
0.040
0.016
0.005
0.000
building
0.043
0.929
0.022
0.005
0.001
0.000
low_veg
0.057
0.022
0.781
0.140
0.000
0.000
tree
0.023
0.005
0.097
0.875
0.000
0.000
car
0.242
0.027
0.007
0.004
0.720
0.000
clutter
0.320
0.292
0.019
0.006
0.021
0.342
Precision/Correctness
0.880
0.932
0.813
0.845
0.835
0.948
Recall/Completeness
0.908
0.929
0.781
0.875
0.720
0.342
F1
0.893
0.931
0.796
0.860
0.773
0.503

Overall accuracy 0.871

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
0.933
0.024
0.031
0.008
0.003
0.000
building
0.035
0.948
0.015
0.002
0.000
0.000
low_veg
0.044
0.014
0.816
0.126
0.000
0.000
tree
0.011
0.002
0.080
0.907
0.000
0.000
car
0.132
0.030
0.004
0.002
0.832
0.001
clutter
0.283
0.308
0.015
0.002
0.021
0.372
Precision/Correctness
0.913
0.951
0.849
0.874
0.875
0.967
Recall/Completeness
0.933
0.948
0.816
0.907
0.832
0.372
F1
0.923
0.949
0.832
0.890
0.853
0.537

Overall accuracy 0.901
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9450
0.0151
0.0216
0.0122
0.0061
0.0000
building
0.0428
0.9323
0.0210
0.0038
0.0002
0.0000
low_veg
0.0703
0.0207
0.7308
0.1782
0.0001
0.0000
tree
0.0292
0.0049
0.0861
0.8774
0.0004
0.0019
car
0.2784
0.0056
0.0010
0.0007
0.7142
0.0000
clutter
0.0000
0.0000
0.0000
0.0114
0.0000
0.9886
Precision/Correctness
0.860
0.955
0.850
0.819
0.868
0.975
Recall/Completeness
0.945
0.932
0.731
0.877
0.714
0.989
F1
0.900
0.943
0.786
0.847
0.784
0.982

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9688
0.0114
0.0115
0.0038
0.0045
0.0000
building
0.0342
0.9523
0.0126
0.0006
0.0002
0.0000
low_veg
0.0514
0.0129
0.7637
0.1719
0.0001
0.0000
tree
0.0137
0.0020
0.0716
0.9115
0.0002
0.0010
car
0.1458
0.0046
0.0003
0.0002
0.8492
0.0000
clutter
0.0000
0.0000
0.0000
0.0026
0.0000
0.9974
Precision/Correctness
0.903
0.972
0.885
0.843
0.893
0.987
Recall/Completeness
0.969
0.952
0.764
0.911
0.849
0.997
F1
0.935
0.962
0.820
0.876
0.870
0.992

Overall accuracy 0.901

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
0.9341
0.0452
0.0103
0.0061
0.0045
0.0000
building
0.0666
0.9216
0.0094
0.0018
0.0005
0.0000
low_veg
0.1337
0.0492
0.5019
0.3137
0.0014
0.0000
tree
0.0433
0.0152
0.0576
0.8832
0.0007
0.0000
car
0.2937
0.0120
0.0005
0.0020
0.6917
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.846
0.970
0.670
0.716
0.855
--
Recall/Completeness
0.934
0.922
0.502
0.883
0.692
--
F1
0.888
0.945
0.574
0.791
0.765
--

Overall accuracy 0.901

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
0.9604
0.0275
0.0078
0.0020
0.0023
0.0000
building
0.0534
0.9373
0.0082
0.0006
0.0005
0.0000
low_veg
0.0936
0.0217
0.5442
0.3398
0.0007
0.0000
tree
0.0182
0.0064
0.0528
0.9220
0.0006
0.0000
car
0.1768
0.0070
0.0001
0.0005
0.8156
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.878
0.984
0.696
0.749
0.900
--
Recall/Completeness
0.960
0.937
0.544
0.922
0.816
--
F1
0.917
0.960
0.611
0.826
0.856
--

Overall accuracy 0.926

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
0.9085
0.0480
0.0300
0.0083
0.0053
0.0000
building
0.0441
0.9366
0.0151
0.0041
0.0002
0.0000
low_veg
0.0959
0.0300
0.7164
0.1575
0.0003
0.0000
tree
0.0193
0.0037
0.0635
0.9132
0.0003
0.0000
car
0.2850
0.0136
0.0003
0.0012
0.6999
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.869
0.949
0.712
0.928
0.883
--
Recall/Completeness
0.908
0.937
0.716
0.913
0.700
--
F1
0.888
0.943
0.714
0.921
0.781
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9406
0.0333
0.0214
0.0023
0.0024
0.0000
building
0.0345
0.9533
0.0108
0.0013
0.0001
0.0000
low_veg
0.0777
0.0137
0.7706
0.1380
0.0000
0.0000
tree
0.0076
0.0012
0.0513
0.9399
0.0001
0.0000
car
0.1684
0.0079
0.0001
0.0000
0.8235
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.908
0.970
0.755
0.951
0.940
--
Recall/Completeness
0.941
0.953
0.771
0.940
0.823
--
F1
0.924
0.962
0.763
0.945
0.878
--

Overall accuracy 0.927

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
0.8896
0.0257
0.0698
0.0103
0.0043
0.0003
building
0.0358
0.9418
0.0175
0.0045
0.0003
0.0000
low_veg
0.0608
0.0124
0.8743
0.0522
0.0003
0.0000
tree
0.0268
0.0058
0.0907
0.8765
0.0001
0.0000
car
0.2634
0.0108
0.0201
0.0033
0.7018
0.0006
clutter
--
--
--
--
--
--
Precision/Correctness
0.900
0.947
0.782
0.935
0.888
--
Recall/Completeness
0.890
0.942
0.874
0.876
0.702
--
F1
0.895
0.945
0.825
0.905
0.784
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9152
0.0192
0.0596
0.0033
0.0024
0.0003
building
0.0255
0.9624
0.0100
0.0020
0.0001
0.0000
low_veg
0.0438
0.0050
0.9095
0.0417
0.0000
0.0000
tree
0.0122
0.0024
0.0697
0.9157
0.0000
0.0000
car
0.1398
0.0079
0.0143
0.0029
0.8344
0.0007
clutter
--
--
--
--
--
--
Precision/Correctness
0.937
0.968
0.823
0.958
0.932
--
Recall/Completeness
0.915
0.962
0.910
0.916
0.834
--
F1
0.926
0.965
0.864
0.936
0.880
--

Overall accuracy 0.924

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
0.0642
0.0454
0.0177
0.0037
0.0008
building
0.0306
0.9446
0.0119
0.0104
0.0024
0.0001
low_veg
0.0891
0.0261
0.7231
0.1615
0.0002
0.0000
tree
0.0194
0.0082
0.0884
0.8840
0.0001
0.0000
car
0.2123
0.0464
0.0142
0.0000
0.7272
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.920
0.879
0.769
0.829
0.719
--
Recall/Completeness
0.868
0.945
0.723
0.884
0.727
--
F1
0.893
0.911
0.745
0.856
0.723
--

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
0.8974
0.0538
0.0380
0.0076
0.0024
0.0009
building
0.0237
0.9592
0.0076
0.0073
0.0021
0.0000
low_veg
0.0846
0.0142
0.7642
0.1368
0.0002
0.0000
tree
0.0087
0.0044
0.0688
0.9180
0.0001
0.0000
car
0.1129
0.0510
0.0140
0.0000
0.8221
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.939
0.905
0.810
0.869
0.747
--
Recall/Completeness
0.897
0.959
0.764
0.918
0.822
--
F1
0.918
0.931
0.786
0.893
0.783
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8686
0.0434
0.0563
0.0294
0.0022
0.0000
building
0.0256
0.9496
0.0183
0.0066
0.0000
0.0000
low_veg
0.0409
0.0223
0.7629
0.1738
0.0001
0.0001
tree
0.0099
0.0053
0.1112
0.8735
0.0000
0.0000
car
0.2217
0.0004
0.0001
0.0022
0.7756
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.851
0.925
0.808
0.853
0.918
--
Recall/Completeness
0.869
0.950
0.763
0.874
0.776
--
F1
0.860
0.937
0.785
0.863
0.841
--

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_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8941
0.0422
0.0477
0.0153
0.0007
0.0000
building
0.0240
0.9668
0.0073
0.0019
0.0000
0.0000
low_veg
0.0371
0.0146
0.7947
0.1535
0.0000
0.0001
tree
0.0046
0.0018
0.0931
0.9004
0.0000
0.0000
car
0.0903
0.0000
0.0000
0.0001
0.9096
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.881
0.945
0.840
0.879
0.970
--
Recall/Completeness
0.894
0.967
0.795
0.900
0.910
--
F1
0.887
0.956
0.817
0.890
0.939
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8517
0.0295
0.0827
0.0308
0.0053
0.0000
building
0.0113
0.9652
0.0183
0.0050
0.0000
0.0002
low_veg
0.0895
0.0502
0.7077
0.1520
0.0006
0.0000
tree
0.0267
0.0104
0.1322
0.8306
0.0001
0.0000
car
0.1648
0.0000
0.0000
0.0048
0.8304
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.851
0.880
0.730
0.856
0.612
--
Recall/Completeness
0.852
0.965
0.708
0.831
0.830
--
F1
0.852
0.921
0.719
0.843
0.705
--

Overall accuracy 0.828

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area14_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8720
0.0247
0.0779
0.0210
0.0044
0.0000
building
0.0064
0.9827
0.0088
0.0018
0.0000
0.0002
low_veg
0.0836
0.0361
0.7368
0.1429
0.0005
0.0000
tree
0.0183
0.0045
0.1193
0.8579
0.0001
0.0000
car
0.0542
0.0000
0.0000
0.0018
0.9440
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.878
0.913
0.753
0.877
0.612
--
Recall/Completeness
0.872
0.983
0.737
0.858
0.944
--
F1
0.875
0.947
0.745
0.867
0.743
--

Overall accuracy 0.853

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
0.8329
0.0273
0.0871
0.0508
0.0019
0.0000
building
0.0214
0.9401
0.0291
0.0090
0.0001
0.0004
low_veg
0.0149
0.0138
0.8392
0.1311
0.0000
0.0010
tree
0.0244
0.0030
0.1157
0.8569
0.0000
0.0000
car
0.1790
0.0006
0.0324
0.0012
0.7868
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.880
0.947
0.803
0.867
0.846
--
Recall/Completeness
0.833
0.940
0.839
0.857
0.787
--
F1
0.856
0.943
0.821
0.862
0.815
--

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
0.8669
0.0228
0.0780
0.0316
0.0007
0.0000
building
0.0175
0.9648
0.0146
0.0027
0.0000
0.0004
low_veg
0.0107
0.0089
0.8620
0.1174
0.0000
0.0011
tree
0.0167
0.0007
0.1008
0.8817
0.0000
0.0000
car
0.1022
0.0000
0.0212
0.0001
0.8764
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.912
0.965
0.833
0.890
0.924
--
Recall/Completeness
0.867
0.965
0.862
0.882
0.876
--
F1
0.889
0.965
0.847
0.886
0.900
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8746
0.0351
0.0368
0.0465
0.0071
0.0000
building
0.0126
0.9693
0.0105
0.0075
0.0000
0.0000
low_veg
0.0528
0.0354
0.7185
0.1932
0.0000
0.0000
tree
0.0089
0.0023
0.0432
0.9455
0.0001
0.0000
car
0.1146
0.0041
0.0188
0.0055
0.8569
0.0000
clutter
0.0038
0.9952
0.0004
0.0006
0.0000
0.0000
Precision/Correctness
0.914
0.791
0.838
0.890
0.709
-nan
Recall/Completeness
0.875
0.969
0.719
0.946
0.857
0.000
F1
0.894
0.871
0.774
0.917
0.776
-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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9073
0.0250
0.0340
0.0282
0.0056
0.0000
building
0.0092
0.9818
0.0068
0.0022
0.0000
0.0000
low_veg
0.0462
0.0246
0.7656
0.1635
0.0000
0.0000
tree
0.0039
0.0007
0.0316
0.9638
0.0000
0.0000
car
0.0319
0.0015
0.0061
0.0002
0.9603
0.0000
clutter
0.0012
0.9987
0.0000
0.0001
0.0000
0.0000
Precision/Correctness
0.939
0.811
0.870
0.917
0.717
-nan
Recall/Completeness
0.907
0.982
0.766
0.964
0.960
0.000
F1
0.923
0.888
0.814
0.940
0.821
-nan

Overall accuracy 0.892

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9021
0.0174
0.0436
0.0308
0.0061
0.0000
building
0.0360
0.9368
0.0193
0.0079
0.0001
0.0000
low_veg
0.0906
0.0256
0.7447
0.1388
0.0003
0.0000
tree
0.0257
0.0039
0.0659
0.9041
0.0004
0.0000
car
0.2490
0.0000
0.0071
0.0137
0.7296
0.0006
clutter
--
--
--
--
--
--
Precision/Correctness
0.892
0.934
0.805
0.869
0.789
--
Recall/Completeness
0.902
0.937
0.745
0.904
0.730
--
F1
0.897
0.935
0.774
0.886
0.758
--

Overall accuracy 0.875

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
0.9318
0.0118
0.0365
0.0162
0.0038
0.0000
building
0.0289
0.9578
0.0098
0.0036
0.0000
0.0000
low_veg
0.0858
0.0157
0.7860
0.1124
0.0001
0.0000
tree
0.0114
0.0012
0.0489
0.9384
0.0002
0.0000
car
0.1510
0.0000
0.0033
0.0064
0.8384
0.0009
clutter
--
--
--
--
--
--
Precision/Correctness
0.919
0.960
0.847
0.906
0.839
--
Recall/Completeness
0.932
0.958
0.786
0.938
0.838
--
F1
0.926
0.959
0.815
0.922
0.839
--

Overall accuracy 0.909

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
0.8348
0.0306
0.0925
0.0354
0.0068
0.0000
building
0.0550
0.8817
0.0529
0.0103
0.0001
0.0000
low_veg
0.0227
0.0206
0.8808
0.0756
0.0003
0.0000
tree
0.0192
0.0078
0.1610
0.8118
0.0003
0.0000
car
0.2324
0.0020
0.0020
0.0015
0.7621
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.827
0.918
0.849
0.844
0.711
--
Recall/Completeness
0.835
0.882
0.881
0.812
0.762
--
F1
0.831
0.899
0.865
0.827
0.736
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8635
0.0219
0.0834
0.0261
0.0052
0.0000
building
0.0519
0.8957
0.0448
0.0076
0.0001
0.0000
low_veg
0.0190
0.0164
0.8991
0.0653
0.0003
0.0000
tree
0.0122
0.0045
0.1386
0.8445
0.0002
0.0000
car
0.1327
0.0009
0.0000
0.0000
0.8664
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.853
0.939
0.872
0.866
0.719
--
Recall/Completeness
0.864
0.896
0.899
0.845
0.866
--
F1
0.858
0.917
0.885
0.855
0.786
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9306
0.0347
0.0209
0.0067
0.0071
0.0000
building
0.0794
0.8970
0.0195
0.0038
0.0003
0.0000
low_veg
0.0817
0.0285
0.6880
0.2008
0.0010
0.0000
tree
0.0148
0.0051
0.0664
0.9135
0.0002
0.0000
car
0.2755
0.0086
0.0055
0.0029
0.7069
0.0006
clutter
0.4579
0.2668
0.1849
0.0132
0.0624
0.0147
Precision/Correctness
0.843
0.945
0.866
0.681
0.893
0.665
Recall/Completeness
0.931
0.897
0.688
0.913
0.707
0.015
F1
0.885
0.921
0.767
0.780
0.789
0.029

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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9546
0.0261
0.0142
0.0025
0.0026
0.0000
building
0.0680
0.9162
0.0143
0.0013
0.0002
0.0000
low_veg
0.0616
0.0163
0.7293
0.1920
0.0008
0.0000
tree
0.0065
0.0011
0.0492
0.9432
0.0000
0.0000
car
0.1582
0.0069
0.0044
0.0021
0.8278
0.0006
clutter
0.4705
0.2646
0.1700
0.0027
0.0713
0.0208
Precision/Correctness
0.879
0.965
0.902
0.710
0.945
0.720
Recall/Completeness
0.955
0.916
0.729
0.943
0.828
0.021
F1
0.915
0.940
0.807
0.810
0.882
0.040

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9387
0.0169
0.0261
0.0123
0.0059
0.0001
building
0.0281
0.9512
0.0141
0.0065
0.0000
0.0000
low_veg
0.0635
0.0164
0.6619
0.2579
0.0003
0.0000
tree
0.0052
0.0025
0.0330
0.9592
0.0000
0.0000
car
0.2317
0.0029
0.0003
0.0074
0.7576
0.0000
clutter
0.3792
0.4439
0.0376
0.0098
0.0397
0.0898
Precision/Correctness
0.926
0.918
0.833
0.697
0.731
0.983
Recall/Completeness
0.939
0.951
0.662
0.959
0.758
0.090
F1
0.932
0.934
0.738
0.808
0.744
0.165

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9608
0.0105
0.0189
0.0054
0.0044
0.0000
building
0.0209
0.9677
0.0081
0.0034
0.0000
0.0000
low_veg
0.0417
0.0063
0.7063
0.2456
0.0002
0.0000
tree
0.0011
0.0005
0.0232
0.9752
0.0000
0.0000
car
0.1345
0.0003
0.0000
0.0044
0.8607
0.0000
clutter
0.3528
0.4743
0.0282
0.0039
0.0382
0.1025
Precision/Correctness
0.948
0.937
0.876
0.737
0.749
0.996
Recall/Completeness
0.961
0.968
0.706
0.975
0.861
0.102
F1
0.955
0.952
0.782
0.839
0.801
0.186

Overall accuracy 0.905

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
0.9704
0.0109
0.0068
0.0047
0.0067
0.0005
building
0.0530
0.9362
0.0075
0.0018
0.0013
0.0001
low_veg
0.0480
0.0047
0.9092
0.0363
0.0010
0.0007
tree
0.0443
0.0054
0.0909
0.8589
0.0005
0.0000
car
0.1875
0.1034
0.0042
0.0039
0.7006
0.0003
clutter
0.5337
0.1728
0.0364
0.0096
0.0308
0.2167
Precision/Correctness
0.880
0.945
0.892
0.923
0.857
0.952
Recall/Completeness
0.970
0.936
0.909
0.859
0.701
0.217
F1
0.923
0.941
0.900
0.890
0.771
0.353

Overall accuracy 0.904

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
0.9832
0.0086
0.0040
0.0012
0.0028
0.0002
building
0.0420
0.9540
0.0027
0.0002
0.0010
0.0001
low_veg
0.0276
0.0026
0.9397
0.0295
0.0004
0.0003
tree
0.0181
0.0014
0.0723
0.9081
0.0000
0.0000
car
0.0973
0.1229
0.0017
0.0015
0.7765
0.0002
clutter
0.4969
0.1879
0.0352
0.0036
0.0325
0.2439
Precision/Correctness
0.914
0.955
0.925
0.951
0.902
0.981
Recall/Completeness
0.983
0.954
0.940
0.908
0.776
0.244
F1
0.947
0.954
0.932
0.929
0.834
0.391

Overall accuracy 0.932

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
0.9633
0.0206
0.0066
0.0040
0.0034
0.0021
building
0.0595
0.8890
0.0455
0.0040
0.0013
0.0008
low_veg
0.0785
0.0119
0.8744
0.0344
0.0008
0.0000
tree
0.0364
0.0065
0.1354
0.8216
0.0000
0.0000
car
0.1672
0.0431
0.0029
0.0006
0.7808
0.0053
clutter
0.4537
0.0744
0.0008
0.0011
0.0220
0.4481
Precision/Correctness
0.861
0.945
0.818
0.915
0.769
0.956
Recall/Completeness
0.963
0.889
0.874
0.822
0.781
0.448
F1
0.909
0.916
0.845
0.866
0.775
0.610

Overall accuracy 0.883

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
0.9765
0.0170
0.0028
0.0013
0.0019
0.0004
building
0.0520
0.9012
0.0430
0.0020
0.0012
0.0006
low_veg
0.0463
0.0048
0.9163
0.0321
0.0005
0.0000
tree
0.0137
0.0024
0.1231
0.8609
0.0000
0.0000
car
0.0643
0.0370
0.0004
0.0001
0.8924
0.0058
clutter
0.4159
0.0718
0.0001
0.0007
0.0235
0.4880
Precision/Correctness
0.892
0.959
0.839
0.936
0.791
0.984
Recall/Completeness
0.977
0.901
0.916
0.861
0.892
0.488
F1
0.932
0.929
0.876
0.897
0.839
0.652

Overall accuracy 0.908

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
0.8703
0.0516
0.0588
0.0182
0.0011
0.0000
building
0.0362
0.9344
0.0226
0.0068
0.0001
0.0000
low_veg
0.0553
0.0141
0.7062
0.2243
0.0001
0.0000
tree
0.0391
0.0081
0.1084
0.8442
0.0002
0.0000
car
0.2723
0.0048
0.0018
0.0086
0.7124
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.863
0.931
0.791
0.761
0.937
--
Recall/Completeness
0.870
0.934
0.706
0.844
0.712
--
F1
0.867
0.933
0.746
0.800
0.810
--

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_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9081
0.0459
0.0407
0.0051
0.0002
0.0000
building
0.0261
0.9604
0.0121
0.0013
0.0000
0.0000
low_veg
0.0388
0.0065
0.7369
0.2177
0.0000
0.0000
tree
0.0154
0.0034
0.0829
0.8983
0.0000
0.0000
car
0.1463
0.0069
0.0010
0.0032
0.8426
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.913
0.952
0.848
0.789
0.985
--
Recall/Completeness
0.908
0.960
0.737
0.898
0.843
--
F1
0.911
0.956
0.789
0.840
0.908
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8927
0.0316
0.0575
0.0141
0.0039
0.0003
building
0.0214
0.9238
0.0472
0.0061
0.0015
0.0001
low_veg
0.0379
0.0157
0.8541
0.0917
0.0005
0.0000
tree
0.0216
0.0040
0.1260
0.8482
0.0002
0.0000
car
0.2298
0.0290
0.0254
0.0049
0.7106
0.0003
clutter
--
--
--
--
--
--
Precision/Correctness
0.914
0.944
0.821
0.842
0.810
--
Recall/Completeness
0.893
0.924
0.854
0.848
0.711
--
F1
0.903
0.934
0.837
0.845
0.757
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9204
0.0299
0.0423
0.0051
0.0020
0.0003
building
0.0141
0.9515
0.0307
0.0024
0.0013
0.0001
low_veg
0.0225
0.0121
0.8889
0.0762
0.0003
0.0000
tree
0.0077
0.0013
0.1011
0.8899
0.0000
0.0000
car
0.1307
0.0337
0.0151
0.0016
0.8187
0.0002
clutter
--
--
--
--
--
--
Precision/Correctness
0.952
0.955
0.864
0.880
0.860
--
Recall/Completeness
0.920
0.952
0.889
0.890
0.819
--
F1
0.936
0.953
0.876
0.885
0.839
--

Overall accuracy 0.913

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image