Detailled semantic labeling (2D) result


Name H. Wang
Affiliation NLPR, Inst. of Automation, China
Abbreviation GSN5
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
0.029
0.042
0.016
0.005
0.000
building
0.048
0.925
0.021
0.005
0.001
0.000
low_veg
0.058
0.021
0.786
0.134
0.000
0.000
tree
0.020
0.005
0.092
0.883
0.000
0.000
car
0.272
0.036
0.010
0.005
0.677
0.000
clutter
0.494
0.189
0.017
0.005
0.031
0.264
Precision/Correctness
0.871
0.937
0.817
0.852
0.818
0.978
Recall/Completeness
0.908
0.925
0.786
0.883
0.677
0.264
F1
0.889
0.931
0.801
0.867
0.741
0.416

Overall accuracy 0.872

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
0.934
0.023
0.034
0.007
0.003
0.000
building
0.039
0.943
0.015
0.002
0.001
0.000
low_veg
0.045
0.013
0.822
0.120
0.000
0.000
tree
0.009
0.002
0.074
0.914
0.000
0.000
car
0.164
0.040
0.007
0.002
0.786
0.000
clutter
0.465
0.196
0.014
0.001
0.032
0.292
Precision/Correctness
0.903
0.956
0.852
0.881
0.855
0.987
Recall/Completeness
0.934
0.943
0.822
0.914
0.786
0.292
F1
0.918
0.950
0.837
0.897
0.819
0.451

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
0.9405
0.0147
0.0278
0.0110
0.0060
0.0000
building
0.0412
0.9358
0.0204
0.0025
0.0001
0.0000
low_veg
0.0712
0.0173
0.7359
0.1753
0.0003
0.0000
tree
0.0256
0.0059
0.0878
0.8792
0.0001
0.0014
car
0.2678
0.0131
0.0005
0.0013
0.7173
0.0000
clutter
0.0027
0.0000
0.0000
0.0100
0.0000
0.9874
Precision/Correctness
0.863
0.957
0.844
0.824
0.873
0.982
Recall/Completeness
0.941
0.936
0.736
0.879
0.717
0.987
F1
0.900
0.946
0.786
0.851
0.787
0.985

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9648
0.0109
0.0178
0.0026
0.0038
0.0000
building
0.0333
0.9547
0.0116
0.0003
0.0001
0.0000
low_veg
0.0537
0.0105
0.7663
0.1693
0.0002
0.0000
tree
0.0110
0.0022
0.0744
0.9119
0.0000
0.0005
car
0.1395
0.0131
0.0000
0.0003
0.8471
0.0000
clutter
0.0018
0.0000
0.0000
0.0015
0.0000
0.9966
Precision/Correctness
0.904
0.975
0.877
0.846
0.908
0.993
Recall/Completeness
0.965
0.955
0.766
0.912
0.847
0.997
F1
0.934
0.965
0.818
0.878
0.876
0.995

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
0.9421
0.0342
0.0149
0.0044
0.0044
0.0000
building
0.0700
0.9192
0.0086
0.0014
0.0008
0.0000
low_veg
0.1411
0.0550
0.5374
0.2654
0.0011
0.0000
tree
0.0410
0.0123
0.0533
0.8929
0.0005
0.0000
car
0.3191
0.0114
0.0004
0.0023
0.6669
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.841
0.975
0.670
0.755
0.842
--
Recall/Completeness
0.942
0.919
0.537
0.893
0.667
--
F1
0.888
0.946
0.597
0.818
0.744
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9685
0.0165
0.0110
0.0012
0.0028
0.0000
building
0.0566
0.9348
0.0074
0.0004
0.0008
0.0000
low_veg
0.1046
0.0338
0.5705
0.2901
0.0010
0.0000
tree
0.0157
0.0057
0.0441
0.9345
0.0000
0.0000
car
0.1938
0.0068
0.0000
0.0012
0.7983
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.872
0.988
0.706
0.783
0.870
--
Recall/Completeness
0.968
0.935
0.570
0.934
0.798
--
F1
0.918
0.961
0.631
0.852
0.832
--

Overall accuracy 0.928

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
0.9072
0.0419
0.0371
0.0085
0.0053
0.0000
building
0.0449
0.9370
0.0145
0.0033
0.0003
0.0000
low_veg
0.0817
0.0202
0.7443
0.1535
0.0004
0.0000
tree
0.0130
0.0026
0.0620
0.9223
0.0001
0.0000
car
0.2982
0.0233
0.0014
0.0014
0.6758
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.878
0.957
0.712
0.931
0.880
--
Recall/Completeness
0.907
0.937
0.744
0.922
0.676
--
F1
0.892
0.947
0.728
0.927
0.765
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9406
0.0276
0.0273
0.0018
0.0027
0.0000
building
0.0354
0.9535
0.0103
0.0006
0.0002
0.0000
low_veg
0.0704
0.0077
0.7890
0.1328
0.0000
0.0000
tree
0.0045
0.0004
0.0495
0.9456
0.0000
0.0000
car
0.1875
0.0187
0.0004
0.0000
0.7935
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.912
0.976
0.754
0.954
0.931
--
Recall/Completeness
0.941
0.954
0.789
0.946
0.793
--
F1
0.926
0.965
0.771
0.950
0.857
--

Overall accuracy 0.930

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
0.8847
0.0298
0.0729
0.0094
0.0032
0.0000
building
0.0437
0.9338
0.0180
0.0040
0.0005
0.0000
low_veg
0.0435
0.0125
0.8555
0.0877
0.0008
0.0000
tree
0.0198
0.0048
0.0716
0.9037
0.0001
0.0000
car
0.2820
0.0123
0.0168
0.0039
0.6849
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.908
0.942
0.791
0.911
0.898
--
Recall/Completeness
0.885
0.934
0.856
0.904
0.685
--
F1
0.896
0.938
0.822
0.907
0.777
--

Overall accuracy 0.891

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
0.9100
0.0243
0.0618
0.0024
0.0015
0.0000
building
0.0328
0.9556
0.0098
0.0013
0.0005
0.0000
low_veg
0.0286
0.0055
0.8870
0.0785
0.0004
0.0000
tree
0.0068
0.0022
0.0500
0.9409
0.0000
0.0000
car
0.1668
0.0132
0.0095
0.0027
0.8078
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.943
0.961
0.836
0.933
0.939
--
Recall/Completeness
0.910
0.956
0.887
0.941
0.808
--
F1
0.926
0.958
0.861
0.937
0.869
--

Overall accuracy 0.923

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
0.8478
0.0656
0.0655
0.0172
0.0038
0.0000
building
0.0350
0.9344
0.0202
0.0094
0.0009
0.0000
low_veg
0.0766
0.0287
0.7606
0.1340
0.0002
0.0000
tree
0.0121
0.0069
0.1001
0.8809
0.0000
0.0000
car
0.2223
0.0412
0.0041
0.0011
0.7313
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.925
0.876
0.728
0.849
0.755
--
Recall/Completeness
0.848
0.934
0.761
0.881
0.731
--
F1
0.885
0.904
0.744
0.865
0.743
--

Overall accuracy 0.859

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
0.8810
0.0531
0.0564
0.0069
0.0026
0.0000
building
0.0297
0.9469
0.0167
0.0060
0.0006
0.0000
low_veg
0.0744
0.0155
0.8021
0.1079
0.0002
0.0000
tree
0.0049
0.0029
0.0766
0.9156
0.0000
0.0000
car
0.1275
0.0451
0.0031
0.0000
0.8243
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.941
0.905
0.767
0.891
0.788
--
Recall/Completeness
0.881
0.947
0.802
0.916
0.824
--
F1
0.910
0.926
0.784
0.903
0.806
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8828
0.0247
0.0593
0.0295
0.0037
0.0000
building
0.0398
0.9443
0.0106
0.0053
0.0000
0.0000
low_veg
0.0484
0.0209
0.7698
0.1604
0.0002
0.0002
tree
0.0105
0.0055
0.0964
0.8876
0.0000
0.0000
car
0.2153
0.0003
0.0000
0.0075
0.7769
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.825
0.940
0.829
0.864
0.867
--
Recall/Completeness
0.883
0.944
0.770
0.888
0.777
--
F1
0.853
0.942
0.798
0.876
0.820
--

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_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9108
0.0224
0.0504
0.0146
0.0019
0.0000
building
0.0372
0.9576
0.0039
0.0014
0.0000
0.0000
low_veg
0.0452
0.0123
0.8052
0.1368
0.0002
0.0002
tree
0.0057
0.0019
0.0796
0.9128
0.0000
0.0000
car
0.0756
0.0000
0.0000
0.0011
0.9233
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.853
0.962
0.859
0.892
0.909
--
Recall/Completeness
0.911
0.958
0.805
0.913
0.923
--
F1
0.881
0.960
0.831
0.902
0.916
--

Overall accuracy 0.891

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
0.8383
0.0299
0.0960
0.0300
0.0058
0.0000
building
0.0127
0.9680
0.0143
0.0050
0.0000
0.0000
low_veg
0.0851
0.0438
0.7233
0.1464
0.0013
0.0000
tree
0.0244
0.0111
0.1190
0.8454
0.0001
0.0000
car
0.1805
0.0000
0.0002
0.0040
0.8153
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.855
0.887
0.741
0.863
0.564
--
Recall/Completeness
0.838
0.968
0.723
0.845
0.815
--
F1
0.847
0.926
0.732
0.854
0.667
--

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_area14_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8589
0.0246
0.0916
0.0198
0.0051
0.0000
building
0.0081
0.9839
0.0070
0.0010
0.0000
0.0000
low_veg
0.0803
0.0290
0.7535
0.1359
0.0014
0.0000
tree
0.0169
0.0048
0.1047
0.8735
0.0000
0.0000
car
0.0631
0.0000
0.0000
0.0000
0.9369
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.880
0.922
0.764
0.884
0.545
--
Recall/Completeness
0.859
0.984
0.753
0.874
0.937
--
F1
0.870
0.952
0.759
0.879
0.689
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8223
0.0258
0.1003
0.0491
0.0025
0.0000
building
0.0388
0.9159
0.0353
0.0096
0.0004
0.0000
low_veg
0.0226
0.0242
0.8379
0.1148
0.0006
0.0000
tree
0.0220
0.0035
0.1140
0.8605
0.0000
0.0000
car
0.1583
0.0017
0.0383
0.0038
0.7979
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.852
0.929
0.796
0.879
0.740
--
Recall/Completeness
0.822
0.916
0.838
0.860
0.798
--
F1
0.837
0.922
0.816
0.870
0.768
--

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_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8566
0.0214
0.0919
0.0292
0.0009
0.0000
building
0.0349
0.9399
0.0221
0.0027
0.0003
0.0000
low_veg
0.0193
0.0199
0.8604
0.0999
0.0005
0.0000
tree
0.0148
0.0011
0.0985
0.8856
0.0000
0.0000
car
0.0986
0.0000
0.0219
0.0000
0.8794
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.880
0.946
0.825
0.904
0.798
--
Recall/Completeness
0.857
0.940
0.860
0.886
0.879
--
F1
0.868
0.943
0.842
0.895
0.837
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8710
0.0269
0.0518
0.0471
0.0032
0.0000
building
0.0103
0.9681
0.0156
0.0059
0.0000
0.0000
low_veg
0.0570
0.0253
0.7358
0.1809
0.0010
0.0000
tree
0.0062
0.0012
0.0480
0.9446
0.0000
0.0000
car
0.1129
0.0199
0.0148
0.0123
0.8401
0.0000
clutter
0.5740
0.4236
0.0014
0.0010
0.0000
0.0000
Precision/Correctness
0.849
0.876
0.813
0.895
0.800
-nan
Recall/Completeness
0.871
0.968
0.736
0.945
0.840
0.000
F1
0.860
0.920
0.772
0.919
0.820
-nan

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9037
0.0201
0.0466
0.0280
0.0014
0.0000
building
0.0072
0.9801
0.0111
0.0016
0.0000
0.0000
low_veg
0.0501
0.0160
0.7782
0.1547
0.0011
0.0000
tree
0.0021
0.0002
0.0354
0.9622
0.0000
0.0000
car
0.0381
0.0237
0.0042
0.0058
0.9282
0.0000
clutter
0.5804
0.4193
0.0000
0.0003
0.0000
0.0000
Precision/Correctness
0.867
0.896
0.847
0.921
0.846
-nan
Recall/Completeness
0.904
0.980
0.778
0.962
0.928
0.000
F1
0.885
0.936
0.811
0.941
0.885
-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
0.8981
0.0243
0.0383
0.0332
0.0061
0.0000
building
0.0334
0.9497
0.0092
0.0072
0.0005
0.0000
low_veg
0.0865
0.0299
0.7366
0.1462
0.0008
0.0000
tree
0.0190
0.0032
0.0566
0.9212
0.0001
0.0000
car
0.2487
0.0140
0.0236
0.0138
0.7000
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.901
0.919
0.830
0.865
0.774
--
Recall/Completeness
0.898
0.950
0.737
0.921
0.700
--
F1
0.899
0.934
0.781
0.892
0.735
--

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_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9296
0.0178
0.0322
0.0164
0.0040
0.0000
building
0.0279
0.9658
0.0029
0.0030
0.0004
0.0000
low_veg
0.0830
0.0176
0.7813
0.1172
0.0008
0.0000
tree
0.0073
0.0008
0.0403
0.9516
0.0000
0.0000
car
0.1351
0.0170
0.0210
0.0045
0.8225
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.925
0.948
0.871
0.905
0.811
--
Recall/Completeness
0.930
0.966
0.781
0.952
0.822
--
F1
0.927
0.957
0.824
0.928
0.817
--

Overall accuracy 0.913

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
0.8379
0.0314
0.0860
0.0381
0.0066
0.0000
building
0.0774
0.8627
0.0444
0.0156
0.0000
0.0000
low_veg
0.0238
0.0117
0.8890
0.0753
0.0002
0.0000
tree
0.0199
0.0061
0.1603
0.8137
0.0000
0.0000
car
0.2044
0.0000
0.0003
0.0031
0.7922
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.802
0.937
0.856
0.840
0.738
--
Recall/Completeness
0.838
0.863
0.889
0.814
0.792
--
F1
0.819
0.898
0.872
0.827
0.764
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8674
0.0220
0.0779
0.0273
0.0053
0.0000
building
0.0723
0.8763
0.0387
0.0127
0.0000
0.0000
low_veg
0.0192
0.0077
0.9073
0.0655
0.0002
0.0000
tree
0.0136
0.0027
0.1374
0.8463
0.0000
0.0000
car
0.0817
0.0000
0.0000
0.0004
0.9178
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.830
0.960
0.878
0.862
0.741
--
Recall/Completeness
0.867
0.876
0.907
0.846
0.918
--
F1
0.848
0.916
0.892
0.854
0.820
--

Overall accuracy 0.881

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
0.9388
0.0276
0.0176
0.0052
0.0107
0.0000
building
0.0784
0.9021
0.0161
0.0031
0.0004
0.0000
low_veg
0.0866
0.0295
0.7042
0.1792
0.0006
0.0000
tree
0.0132
0.0045
0.0764
0.9058
0.0002
0.0000
car
0.3093
0.0036
0.0050
0.0051
0.6770
0.0000
clutter
0.5561
0.2055
0.1132
0.0069
0.1183
0.0000
Precision/Correctness
0.840
0.953
0.877
0.705
0.847
-nan
Recall/Completeness
0.939
0.902
0.704
0.906
0.677
0.000
F1
0.887
0.927
0.781
0.793
0.753
-nan

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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9648
0.0182
0.0108
0.0018
0.0044
0.0000
building
0.0660
0.9219
0.0109
0.0009
0.0003
0.0000
low_veg
0.0655
0.0163
0.7488
0.1691
0.0002
0.0000
tree
0.0053
0.0011
0.0584
0.9352
0.0000
0.0000
car
0.2120
0.0018
0.0029
0.0044
0.7788
0.0000
clutter
0.5850
0.1846
0.0930
0.0038
0.1337
0.0000
Precision/Correctness
0.876
0.973
0.914
0.734
0.914
-nan
Recall/Completeness
0.965
0.922
0.749
0.935
0.779
0.000
F1
0.918
0.947
0.823
0.823
0.841
-nan

Overall accuracy 0.899

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
0.9427
0.0172
0.0214
0.0133
0.0054
0.0001
building
0.0231
0.9574
0.0131
0.0064
0.0001
0.0000
low_veg
0.0473
0.0176
0.6655
0.2693
0.0004
0.0000
tree
0.0046
0.0015
0.0237
0.9701
0.0000
0.0001
car
0.2896
0.0026
0.0069
0.0119
0.6891
0.0000
clutter
0.5445
0.3673
0.0282
0.0060
0.0501
0.0039
Precision/Correctness
0.922
0.925
0.859
0.691
0.706
0.740
Recall/Completeness
0.943
0.957
0.665
0.970
0.689
0.004
F1
0.932
0.941
0.750
0.807
0.697
0.008

Overall accuracy 0.877

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
0.9652
0.0107
0.0144
0.0056
0.0042
0.0000
building
0.0167
0.9725
0.0078
0.0029
0.0001
0.0000
low_veg
0.0273
0.0068
0.7099
0.2558
0.0003
0.0000
tree
0.0016
0.0001
0.0160
0.9822
0.0000
0.0000
car
0.1881
0.0009
0.0068
0.0090
0.7952
0.0000
clutter
0.5301
0.3884
0.0190
0.0026
0.0554
0.0045
Precision/Correctness
0.944
0.945
0.901
0.731
0.708
0.896
Recall/Completeness
0.965
0.973
0.710
0.982
0.795
0.004
F1
0.954
0.958
0.794
0.838
0.749
0.009

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9749
0.0057
0.0107
0.0036
0.0051
0.0001
building
0.0596
0.9247
0.0092
0.0024
0.0039
0.0002
low_veg
0.0386
0.0027
0.9146
0.0433
0.0005
0.0002
tree
0.0321
0.0024
0.0682
0.8968
0.0006
0.0000
car
0.2022
0.1771
0.0060
0.0034
0.6113
0.0000
clutter
0.5593
0.1851
0.0432
0.0079
0.0532
0.1513
Precision/Correctness
0.879
0.943
0.894
0.920
0.815
0.981
Recall/Completeness
0.975
0.925
0.915
0.897
0.611
0.151
F1
0.925
0.934
0.904
0.908
0.699
0.262

Overall accuracy 0.902

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9848
0.0038
0.0083
0.0006
0.0025
0.0000
building
0.0460
0.9455
0.0046
0.0002
0.0035
0.0002
low_veg
0.0204
0.0009
0.9437
0.0350
0.0001
0.0000
tree
0.0085
0.0004
0.0510
0.9400
0.0001
0.0000
car
0.1099
0.1976
0.0017
0.0008
0.6901
0.0000
clutter
0.5147
0.2064
0.0448
0.0019
0.0556
0.1766
Precision/Correctness
0.914
0.953
0.924
0.948
0.843
0.990
Recall/Completeness
0.985
0.945
0.944
0.940
0.690
0.177
F1
0.948
0.949
0.934
0.944
0.759
0.300

Overall accuracy 0.929

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
0.9630
0.0246
0.0088
0.0025
0.0010
0.0002
building
0.0657
0.8516
0.0749
0.0052
0.0024
0.0001
low_veg
0.0552
0.0123
0.8889
0.0432
0.0004
0.0000
tree
0.0392
0.0218
0.1128
0.8262
0.0000
0.0000
car
0.3332
0.0008
0.0025
0.0019
0.6616
0.0000
clutter
0.6412
0.0527
0.0010
0.0012
0.0296
0.2742
Precision/Correctness
0.840
0.937
0.787
0.904
0.756
0.993
Recall/Completeness
0.963
0.852
0.889
0.826
0.662
0.274
F1
0.897
0.892
0.835
0.863
0.706
0.430

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_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9743
0.0212
0.0037
0.0005
0.0003
0.0000
building
0.0583
0.8633
0.0733
0.0027
0.0024
0.0001
low_veg
0.0286
0.0058
0.9259
0.0396
0.0002
0.0000
tree
0.0150
0.0201
0.0985
0.8665
0.0000
0.0000
car
0.1970
0.0001
0.0010
0.0004
0.8015
0.0000
clutter
0.6045
0.0535
0.0002
0.0001
0.0325
0.3092
Precision/Correctness
0.871
0.950
0.807
0.927
0.762
0.999
Recall/Completeness
0.974
0.863
0.926
0.866
0.801
0.309
F1
0.920
0.904
0.862
0.896
0.781
0.472

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8510
0.0679
0.0620
0.0169
0.0021
0.0000
building
0.0353
0.9430
0.0166
0.0049
0.0002
0.0000
low_veg
0.0545
0.0092
0.7203
0.2159
0.0001
0.0000
tree
0.0287
0.0053
0.1004
0.8655
0.0001
0.0000
car
0.2968
0.0062
0.0010
0.0063
0.6898
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.870
0.924
0.804
0.774
0.893
--
Recall/Completeness
0.851
0.943
0.720
0.865
0.690
--
F1
0.860
0.933
0.760
0.817
0.779
--

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_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8887
0.0647
0.0415
0.0043
0.0009
0.0000
building
0.0272
0.9648
0.0073
0.0006
0.0002
0.0000
low_veg
0.0395
0.0024
0.7506
0.2074
0.0001
0.0000
tree
0.0081
0.0011
0.0750
0.9159
0.0000
0.0000
car
0.1808
0.0053
0.0008
0.0008
0.8123
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.915
0.942
0.862
0.801
0.944
--
Recall/Completeness
0.889
0.965
0.751
0.916
0.812
--
F1
0.902
0.953
0.802
0.855
0.873
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9195
0.0232
0.0433
0.0103
0.0032
0.0004
building
0.0377
0.9224
0.0331
0.0053
0.0015
0.0000
low_veg
0.0590
0.0221
0.8404
0.0781
0.0004
0.0000
tree
0.0347
0.0030
0.1221
0.8402
0.0000
0.0000
car
0.3264
0.0019
0.0532
0.0062
0.6123
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.873
0.947
0.841
0.863
0.814
--
Recall/Completeness
0.920
0.922
0.840
0.840
0.612
--
F1
0.896
0.935
0.841
0.851
0.699
--

Overall accuracy 0.880

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9425
0.0207
0.0313
0.0034
0.0016
0.0005
building
0.0283
0.9474
0.0211
0.0020
0.0012
0.0000
low_veg
0.0398
0.0176
0.8794
0.0631
0.0001
0.0000
tree
0.0144
0.0006
0.0991
0.8859
0.0000
0.0000
car
0.2280
0.0000
0.0527
0.0019
0.7174
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.917
0.960
0.880
0.899
0.870
--
Recall/Completeness
0.942
0.947
0.879
0.886
0.717
--
F1
0.930
0.953
0.880
0.892
0.786
--

Overall accuracy 0.914

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image