Detailled semantic labeling (2D) result


Name H. Wang
Affiliation NLPR, Inst. of Automation, China
Abbreviation GSN
Strategy (u)nsupervised, (s)upervised, (h)ybrid s


Overall statistics, reference set: full_reference


Number of non processed tiles by this participant: 1
↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.907
0.032
0.038
0.014
0.008
0.000
building
0.045
0.925
0.025
0.004
0.000
0.000
low_veg
0.056
0.023
0.788
0.132
0.001
0.000
tree
0.022
0.006
0.100
0.872
0.000
0.000
car
0.187
0.027
0.005
0.004
0.777
0.000
clutter
0.286
0.675
0.009
0.006
0.023
0.000
Precision/Correctness
0.882
0.916
0.803
0.860
0.785
-nan
Recall/Completeness
0.907
0.925
0.788
0.872
0.777
0.000
F1
0.895
0.921
0.795
0.866
0.781
-nan

Overall accuracy 0.868

Overall statistics, reference set: no_boundary


Number of non processed tiles by this participant: 1
↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.936
0.023
0.031
0.007
0.004
0.000
building
0.037
0.942
0.019
0.002
0.000
0.000
low_veg
0.044
0.014
0.823
0.119
0.000
0.000
tree
0.010
0.002
0.082
0.906
0.000
0.000
car
0.109
0.030
0.003
0.002
0.856
0.000
clutter
0.249
0.720
0.006
0.002
0.024
0.000
Precision/Correctness
0.914
0.938
0.838
0.886
0.849
-nan
Recall/Completeness
0.936
0.942
0.823
0.906
0.856
0.000
F1
0.924
0.940
0.830
0.896
0.853
-nan

Overall accuracy 0.898
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9461
0.0131
0.0212
0.0091
0.0105
0.0000
building
0.0419
0.9332
0.0221
0.0028
0.0000
0.0000
low_veg
0.0814
0.0192
0.7347
0.1646
0.0001
0.0000
tree
0.0295
0.0064
0.1053
0.8584
0.0004
0.0000
car
0.1787
0.0028
0.0011
0.0008
0.8167
0.0000
clutter
0.0000
0.9885
0.0000
0.0115
0.0000
0.0000
Precision/Correctness
0.857
0.884
0.832
0.830
0.822
-nan
Recall/Completeness
0.946
0.933
0.735
0.858
0.817
0.000
F1
0.899
0.908
0.780
0.844
0.819
-nan

Overall accuracy 0.851

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9705
0.0095
0.0118
0.0022
0.0059
0.0000
building
0.0337
0.9519
0.0139
0.0005
0.0000
0.0000
low_veg
0.0625
0.0099
0.7674
0.1602
0.0000
0.0000
tree
0.0139
0.0022
0.0899
0.8939
0.0002
0.0000
car
0.0801
0.0010
0.0004
0.0001
0.9183
0.0000
clutter
0.0000
0.9978
0.0000
0.0022
0.0000
0.0000
Precision/Correctness
0.897
0.900
0.865
0.851
0.879
-nan
Recall/Completeness
0.971
0.952
0.767
0.894
0.918
0.000
F1
0.932
0.925
0.813
0.872
0.898
-nan

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9112
0.0665
0.0134
0.0030
0.0059
0.0000
building
0.0525
0.9373
0.0090
0.0009
0.0003
0.0000
low_veg
0.1241
0.0575
0.5861
0.2318
0.0005
0.0000
tree
0.0501
0.0148
0.0839
0.8506
0.0005
0.0000
car
0.2607
0.0215
0.0044
0.0016
0.7117
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.866
0.958
0.667
0.776
0.842
--
Recall/Completeness
0.911
0.937
0.586
0.851
0.712
--
F1
0.888
0.948
0.624
0.812
0.771
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9453
0.0414
0.0101
0.0008
0.0025
0.0000
building
0.0411
0.9509
0.0075
0.0002
0.0003
0.0000
low_veg
0.0861
0.0252
0.6302
0.2584
0.0001
0.0000
tree
0.0214
0.0055
0.0737
0.8992
0.0001
0.0000
car
0.1747
0.0150
0.0044
0.0006
0.8052
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.898
0.977
0.700
0.797
0.909
--
Recall/Completeness
0.945
0.951
0.630
0.899
0.805
--
F1
0.921
0.964
0.663
0.845
0.854
--

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_area6_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8871
0.0608
0.0371
0.0067
0.0083
0.0000
building
0.0391
0.9359
0.0221
0.0022
0.0007
0.0000
low_veg
0.0799
0.0257
0.7456
0.1479
0.0009
0.0000
tree
0.0169
0.0036
0.0723
0.9071
0.0002
0.0000
car
0.2450
0.0608
0.0011
0.0005
0.6926
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.882
0.939
0.679
0.935
0.819
--
Recall/Completeness
0.887
0.936
0.746
0.907
0.693
--
F1
0.884
0.937
0.711
0.921
0.751
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9262
0.0416
0.0273
0.0018
0.0031
0.0000
building
0.0314
0.9518
0.0160
0.0004
0.0004
0.0000
low_veg
0.0628
0.0113
0.7892
0.1366
0.0002
0.0000
tree
0.0056
0.0008
0.0560
0.9376
0.0000
0.0000
car
0.1623
0.0565
0.0004
0.0000
0.7808
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.918
0.963
0.726
0.953
0.910
--
Recall/Completeness
0.926
0.952
0.789
0.938
0.781
--
F1
0.922
0.958
0.756
0.945
0.841
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8833
0.0321
0.0700
0.0069
0.0077
0.0000
building
0.0372
0.9356
0.0244
0.0026
0.0002
0.0000
low_veg
0.0471
0.0148
0.8712
0.0658
0.0011
0.0000
tree
0.0271
0.0057
0.0956
0.8713
0.0003
0.0000
car
0.1940
0.0036
0.0138
0.0020
0.7866
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.909
0.937
0.772
0.930
0.829
--
Recall/Completeness
0.883
0.936
0.871
0.871
0.787
--
F1
0.896
0.937
0.818
0.900
0.807
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9102
0.0251
0.0602
0.0013
0.0032
0.0000
building
0.0270
0.9573
0.0149
0.0006
0.0002
0.0000
low_veg
0.0302
0.0074
0.9033
0.0589
0.0003
0.0000
tree
0.0105
0.0027
0.0723
0.9143
0.0001
0.0000
car
0.1135
0.0016
0.0083
0.0013
0.8753
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.946
0.958
0.815
0.948
0.909
--
Recall/Completeness
0.910
0.957
0.903
0.914
0.875
--
F1
0.928
0.958
0.857
0.931
0.892
--

Overall accuracy 0.921

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
0.8767
0.0514
0.0509
0.0163
0.0048
0.0000
building
0.0374
0.9304
0.0223
0.0085
0.0014
0.0000
low_veg
0.0833
0.0423
0.7446
0.1295
0.0004
0.0000
tree
0.0166
0.0074
0.1159
0.8600
0.0001
0.0000
car
0.1251
0.0286
0.0021
0.0020
0.8422
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.922
0.886
0.731
0.851
0.730
--
Recall/Completeness
0.877
0.930
0.745
0.860
0.842
--
F1
0.899
0.907
0.738
0.856
0.782
--

Overall accuracy 0.862

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
0.9102
0.0387
0.0418
0.0068
0.0026
0.0000
building
0.0304
0.9453
0.0177
0.0055
0.0011
0.0000
low_veg
0.0799
0.0311
0.7810
0.1075
0.0004
0.0000
tree
0.0065
0.0032
0.0914
0.8989
0.0000
0.0000
car
0.0603
0.0345
0.0000
0.0003
0.9050
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.940
0.916
0.774
0.891
0.782
--
Recall/Completeness
0.910
0.945
0.781
0.899
0.905
--
F1
0.925
0.930
0.777
0.895
0.839
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8956
0.0262
0.0474
0.0275
0.0034
0.0000
building
0.1004
0.8705
0.0213
0.0079
0.0000
0.0000
low_veg
0.0299
0.0167
0.7929
0.1604
0.0001
0.0000
tree
0.0092
0.0058
0.1047
0.8803
0.0000
0.0000
car
0.1827
0.0054
0.0000
0.0070
0.8049
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.798
0.939
0.823
0.862
0.887
--
Recall/Completeness
0.896
0.870
0.793
0.880
0.805
--
F1
0.844
0.904
0.808
0.871
0.844
--

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
0.9234
0.0238
0.0383
0.0135
0.0010
0.0000
building
0.1037
0.8836
0.0095
0.0032
0.0000
0.0000
low_veg
0.0259
0.0097
0.8258
0.1386
0.0000
0.0000
tree
0.0038
0.0021
0.0860
0.9080
0.0000
0.0000
car
0.0706
0.0043
0.0000
0.0044
0.9207
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.819
0.961
0.856
0.890
0.960
--
Recall/Completeness
0.923
0.884
0.826
0.908
0.921
--
F1
0.868
0.921
0.841
0.899
0.940
--

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
0.8717
0.0272
0.0657
0.0274
0.0080
0.0000
building
0.0188
0.9573
0.0186
0.0053
0.0000
0.0000
low_veg
0.0774
0.0488
0.7589
0.1142
0.0008
0.0000
tree
0.0262
0.0123
0.1424
0.8190
0.0001
0.0000
car
0.1008
0.0000
0.0002
0.0009
0.8980
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.862
0.881
0.745
0.883
0.536
--
Recall/Completeness
0.872
0.957
0.759
0.819
0.898
--
F1
0.867
0.917
0.752
0.850
0.671
--

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_area14_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8911
0.0228
0.0610
0.0186
0.0066
0.0000
building
0.0133
0.9761
0.0085
0.0021
0.0000
0.0000
low_veg
0.0712
0.0344
0.7884
0.1054
0.0007
0.0000
tree
0.0175
0.0055
0.1275
0.8495
0.0001
0.0000
car
0.0281
0.0000
0.0000
0.0000
0.9719
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.889
0.915
0.769
0.903
0.524
--
Recall/Completeness
0.891
0.976
0.788
0.850
0.972
--
F1
0.890
0.945
0.779
0.875
0.681
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8283
0.0204
0.0985
0.0488
0.0039
0.0000
building
0.0363
0.9147
0.0354
0.0134
0.0001
0.0000
low_veg
0.0151
0.0119
0.8617
0.1113
0.0000
0.0000
tree
0.0211
0.0041
0.1147
0.8602
0.0000
0.0000
car
0.1374
0.0024
0.0061
0.0048
0.8494
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.871
0.952
0.801
0.880
0.744
--
Recall/Completeness
0.828
0.915
0.862
0.860
0.849
--
F1
0.849
0.933
0.830
0.870
0.793
--

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_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8612
0.0165
0.0917
0.0288
0.0018
0.0000
building
0.0333
0.9402
0.0208
0.0057
0.0000
0.0000
low_veg
0.0112
0.0072
0.8842
0.0974
0.0000
0.0000
tree
0.0134
0.0014
0.0989
0.8863
0.0000
0.0000
car
0.0930
0.0000
0.0023
0.0008
0.9038
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.900
0.970
0.830
0.905
0.832
--
Recall/Completeness
0.861
0.940
0.884
0.886
0.904
--
F1
0.880
0.955
0.856
0.895
0.867
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8590
0.0458
0.0491
0.0404
0.0056
0.0000
building
0.0079
0.9685
0.0185
0.0052
0.0000
0.0000
low_veg
0.0489
0.0293
0.7502
0.1706
0.0010
0.0000
tree
0.0090
0.0023
0.0523
0.9365
0.0000
0.0000
car
0.0492
0.0013
0.0134
0.0106
0.9255
0.0000
clutter
0.0001
0.9968
0.0027
0.0005
0.0000
0.0000
Precision/Correctness
0.922
0.787
0.808
0.901
0.739
-nan
Recall/Completeness
0.859
0.968
0.750
0.936
0.926
0.000
F1
0.889
0.869
0.778
0.919
0.822
-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
0.8926
0.0361
0.0452
0.0235
0.0026
0.0000
building
0.0059
0.9800
0.0127
0.0013
0.0000
0.0000
low_veg
0.0434
0.0176
0.7927
0.1453
0.0011
0.0000
tree
0.0039
0.0006
0.0387
0.9568
0.0000
0.0000
car
0.0102
0.0001
0.0034
0.0041
0.9823
0.0000
clutter
0.0000
0.9989
0.0011
0.0000
0.0000
0.0000
Precision/Correctness
0.944
0.807
0.842
0.926
0.800
-nan
Recall/Completeness
0.893
0.980
0.793
0.957
0.982
0.000
F1
0.918
0.885
0.817
0.941
0.882
-nan

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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9098
0.0215
0.0310
0.0287
0.0089
0.0000
building
0.0307
0.9507
0.0128
0.0056
0.0002
0.0000
low_veg
0.0820
0.0414
0.7452
0.1313
0.0002
0.0000
tree
0.0242
0.0038
0.0626
0.9090
0.0004
0.0000
car
0.1633
0.0000
0.0214
0.0098
0.8056
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.904
0.912
0.832
0.877
0.744
--
Recall/Completeness
0.910
0.951
0.745
0.909
0.806
--
F1
0.907
0.931
0.786
0.893
0.773
--

Overall accuracy 0.882

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9413
0.0141
0.0249
0.0146
0.0051
0.0000
building
0.0230
0.9691
0.0058
0.0021
0.0000
0.0000
low_veg
0.0770
0.0295
0.7896
0.1038
0.0001
0.0000
tree
0.0105
0.0010
0.0453
0.9430
0.0002
0.0000
car
0.0865
0.0000
0.0214
0.0035
0.8886
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.930
0.943
0.874
0.914
0.806
--
Recall/Completeness
0.941
0.969
0.790
0.943
0.889
--
F1
0.936
0.956
0.829
0.928
0.845
--

Overall accuracy 0.917

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
0.8067
0.0276
0.1306
0.0245
0.0106
0.0000
building
0.0317
0.8680
0.0912
0.0091
0.0000
0.0000
low_veg
0.0170
0.0126
0.8954
0.0747
0.0002
0.0000
tree
0.0169
0.0062
0.1609
0.8159
0.0001
0.0000
car
0.1074
0.0000
0.0023
0.0016
0.8887
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.869
0.938
0.827
0.852
0.671
--
Recall/Completeness
0.807
0.868
0.895
0.816
0.889
--
F1
0.837
0.902
0.860
0.834
0.765
--

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
0.8401
0.0184
0.1208
0.0143
0.0064
0.0000
building
0.0275
0.8812
0.0842
0.0071
0.0000
0.0000
low_veg
0.0139
0.0083
0.9136
0.0639
0.0002
0.0000
tree
0.0104
0.0026
0.1384
0.8487
0.0000
0.0000
car
0.0557
0.0000
0.0000
0.0000
0.9443
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.897
0.961
0.849
0.875
0.719
--
Recall/Completeness
0.840
0.881
0.914
0.849
0.944
--
F1
0.868
0.920
0.880
0.862
0.816
--

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
0.9234
0.0306
0.0234
0.0048
0.0178
0.0000
building
0.0631
0.9102
0.0244
0.0021
0.0003
0.0000
low_veg
0.0852
0.0327
0.7111
0.1703
0.0007
0.0000
tree
0.0149
0.0053
0.1054
0.8741
0.0002
0.0000
car
0.2233
0.0058
0.0022
0.0062
0.7626
0.0000
clutter
0.5241
0.3990
0.0627
0.0044
0.0098
0.0000
Precision/Correctness
0.858
0.947
0.838
0.710
0.806
-nan
Recall/Completeness
0.923
0.910
0.711
0.874
0.763
0.000
F1
0.890
0.928
0.770
0.784
0.784
-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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9547
0.0201
0.0161
0.0016
0.0075
0.0000
building
0.0509
0.9297
0.0187
0.0005
0.0002
0.0000
low_veg
0.0655
0.0185
0.7539
0.1619
0.0003
0.0000
tree
0.0058
0.0018
0.0846
0.9078
0.0000
0.0000
car
0.1407
0.0046
0.0007
0.0048
0.8492
0.0000
clutter
0.5306
0.4165
0.0395
0.0019
0.0115
0.0000
Precision/Correctness
0.894
0.969
0.875
0.738
0.892
-nan
Recall/Completeness
0.955
0.930
0.754
0.908
0.849
0.000
F1
0.923
0.949
0.810
0.814
0.870
-nan

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9461
0.0191
0.0162
0.0107
0.0078
0.0000
building
0.0223
0.9577
0.0142
0.0057
0.0001
0.0000
low_veg
0.0741
0.0195
0.6765
0.2287
0.0012
0.0000
tree
0.0057
0.0021
0.0396
0.9526
0.0000
0.0000
car
0.1702
0.0411
0.0010
0.0083
0.7795
0.0000
clutter
0.2902
0.6390
0.0277
0.0060
0.0372
0.0000
Precision/Correctness
0.934
0.894
0.863
0.722
0.693
-nan
Recall/Completeness
0.946
0.958
0.677
0.953
0.779
0.000
F1
0.940
0.925
0.758
0.821
0.734
-nan

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9685
0.0114
0.0102
0.0046
0.0052
0.0000
building
0.0157
0.9736
0.0079
0.0027
0.0001
0.0000
low_veg
0.0503
0.0081
0.7245
0.2162
0.0009
0.0000
tree
0.0015
0.0003
0.0273
0.9708
0.0000
0.0000
car
0.0889
0.0476
0.0002
0.0054
0.8579
0.0000
clutter
0.2526
0.6864
0.0185
0.0027
0.0398
0.0000
Precision/Correctness
0.956
0.916
0.908
0.761
0.716
-nan
Recall/Completeness
0.969
0.974
0.725
0.971
0.858
0.000
F1
0.962
0.944
0.806
0.853
0.780
-nan

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9700
0.0076
0.0062
0.0040
0.0122
0.0000
building
0.0504
0.9369
0.0078
0.0033
0.0016
0.0000
low_veg
0.0508
0.0052
0.8964
0.0464
0.0013
0.0000
tree
0.0461
0.0031
0.0548
0.8937
0.0022
0.0000
car
0.1455
0.0893
0.0053
0.0026
0.7573
0.0000
clutter
0.5215
0.4156
0.0112
0.0119
0.0398
0.0000
Precision/Correctness
0.884
0.924
0.922
0.911
0.796
-nan
Recall/Completeness
0.970
0.937
0.896
0.894
0.757
0.000
F1
0.925
0.930
0.909
0.902
0.776
-nan

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9857
0.0053
0.0032
0.0012
0.0045
0.0000
building
0.0369
0.9568
0.0043
0.0005
0.0014
0.0000
low_veg
0.0325
0.0016
0.9261
0.0395
0.0003
0.0000
tree
0.0168
0.0004
0.0400
0.9422
0.0006
0.0000
car
0.0859
0.1072
0.0012
0.0004
0.8053
0.0000
clutter
0.4797
0.4665
0.0068
0.0040
0.0430
0.0000
Precision/Correctness
0.918
0.934
0.951
0.939
0.866
-nan
Recall/Completeness
0.986
0.957
0.926
0.942
0.805
0.000
F1
0.951
0.945
0.938
0.941
0.834
-nan

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
0.9642
0.0188
0.0110
0.0022
0.0038
0.0000
building
0.0779
0.8345
0.0823
0.0038
0.0015
0.0000
low_veg
0.0496
0.0165
0.8974
0.0357
0.0007
0.0000
tree
0.0425
0.0087
0.1265
0.8223
0.0001
0.0000
car
0.1804
0.0077
0.0023
0.0012
0.8084
0.0000
clutter
0.3995
0.5703
0.0011
0.0010
0.0281
0.0000
Precision/Correctness
0.864
0.853
0.769
0.920
0.747
-nan
Recall/Completeness
0.964
0.834
0.897
0.822
0.808
0.000
F1
0.911
0.844
0.828
0.868
0.777
-nan

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_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9814
0.0110
0.0058
0.0004
0.0014
0.0000
building
0.0710
0.8449
0.0810
0.0017
0.0015
0.0000
low_veg
0.0225
0.0092
0.9344
0.0334
0.0005
0.0000
tree
0.0165
0.0041
0.1161
0.8633
0.0000
0.0000
car
0.1042
0.0063
0.0006
0.0005
0.8883
0.0000
clutter
0.3666
0.6030
0.0009
0.0000
0.0294
0.0000
Precision/Correctness
0.892
0.875
0.786
0.939
0.777
-nan
Recall/Completeness
0.981
0.845
0.934
0.863
0.888
0.000
F1
0.934
0.860
0.854
0.900
0.829
-nan

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8629
0.0596
0.0567
0.0171
0.0036
0.0000
building
0.0326
0.9492
0.0134
0.0045
0.0003
0.0000
low_veg
0.0612
0.0130
0.7249
0.2004
0.0005
0.0000
tree
0.0351
0.0064
0.1219
0.8363
0.0003
0.0000
car
0.2126
0.0032
0.0000
0.0061
0.7780
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.865
0.928
0.795
0.780
0.832
--
Recall/Completeness
0.863
0.949
0.725
0.836
0.778
--
F1
0.864
0.938
0.758
0.807
0.804
--

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
0.9000
0.0545
0.0388
0.0057
0.0010
0.0000
building
0.0219
0.9715
0.0055
0.0008
0.0003
0.0000
low_veg
0.0435
0.0043
0.7564
0.1953
0.0005
0.0000
tree
0.0110
0.0013
0.0956
0.8920
0.0001
0.0000
car
0.1360
0.0025
0.0000
0.0013
0.8602
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.917
0.949
0.849
0.805
0.914
--
Recall/Completeness
0.900
0.971
0.756
0.892
0.860
--
F1
0.908
0.960
0.800
0.846
0.886
--

Overall accuracy 0.883

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image