Detailled semantic labeling (2D) result


Name H. Wang
Affiliation NLPR, Inst. of Automation, China
Abbreviation GSN4
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
0.909
0.030
0.041
0.016
0.004
0.000
building
0.048
0.926
0.020
0.005
0.001
0.000
low_veg
0.059
0.023
0.776
0.142
0.000
0.000
tree
0.020
0.005
0.085
0.889
0.000
0.000
car
0.297
0.037
0.010
0.005
0.651
0.000
clutter
0.493
0.189
0.018
0.006
0.030
0.265
Precision/Correctness
0.869
0.935
0.824
0.846
0.829
0.979
Recall/Completeness
0.909
0.926
0.776
0.889
0.651
0.265
F1
0.889
0.931
0.799
0.867
0.730
0.417

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
0.934
0.023
0.033
0.008
0.002
0.000
building
0.039
0.944
0.014
0.002
0.001
0.000
low_veg
0.046
0.015
0.813
0.127
0.000
0.000
tree
0.009
0.002
0.069
0.920
0.000
0.000
car
0.182
0.041
0.008
0.002
0.767
0.000
clutter
0.464
0.196
0.015
0.001
0.031
0.293
Precision/Correctness
0.902
0.954
0.857
0.876
0.862
0.988
Recall/Completeness
0.934
0.944
0.813
0.920
0.767
0.293
F1
0.918
0.949
0.834
0.897
0.812
0.452

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
0.9425
0.0152
0.0253
0.0114
0.0055
0.0000
building
0.0414
0.9372
0.0188
0.0025
0.0001
0.0000
low_veg
0.0733
0.0177
0.7256
0.1832
0.0003
0.0000
tree
0.0265
0.0064
0.0812
0.8845
0.0001
0.0012
car
0.2920
0.0084
0.0005
0.0014
0.6976
0.0000
clutter
0.0038
0.0000
0.0000
0.0110
0.0000
0.9852
Precision/Correctness
0.860
0.956
0.852
0.818
0.878
0.984
Recall/Completeness
0.943
0.937
0.726
0.884
0.698
0.985
F1
0.899
0.947
0.784
0.850
0.778
0.984

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
0.9662
0.0114
0.0159
0.0028
0.0036
0.0000
building
0.0334
0.9554
0.0108
0.0003
0.0001
0.0000
low_veg
0.0555
0.0101
0.7572
0.1769
0.0002
0.0000
tree
0.0117
0.0025
0.0688
0.9166
0.0000
0.0005
car
0.1574
0.0075
0.0002
0.0004
0.8346
0.0000
clutter
0.0028
0.0000
0.0000
0.0017
0.0000
0.9954
Precision/Correctness
0.901
0.974
0.885
0.841
0.910
0.994
Recall/Completeness
0.966
0.955
0.757
0.917
0.835
0.995
F1
0.933
0.965
0.816
0.877
0.871
0.995

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9419
0.0350
0.0146
0.0046
0.0039
0.0000
building
0.0692
0.9201
0.0083
0.0016
0.0008
0.0000
low_veg
0.1423
0.0565
0.5246
0.2753
0.0012
0.0000
tree
0.0411
0.0126
0.0439
0.9020
0.0004
0.0000
car
0.3445
0.0118
0.0007
0.0024
0.6405
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.840
0.975
0.680
0.748
0.849
--
Recall/Completeness
0.942
0.920
0.525
0.902
0.641
--
F1
0.888
0.947
0.592
0.818
0.730
--

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
0.0168
0.0108
0.0013
0.0026
0.0000
building
0.0561
0.9354
0.0071
0.0006
0.0008
0.0000
low_veg
0.1060
0.0348
0.5579
0.3002
0.0012
0.0000
tree
0.0156
0.0061
0.0363
0.9420
0.0000
0.0000
car
0.2150
0.0071
0.0000
0.0013
0.7766
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.872
0.988
0.714
0.776
0.872
--
Recall/Completeness
0.968
0.935
0.558
0.942
0.777
--
F1
0.918
0.961
0.626
0.851
0.822
--

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
0.9091
0.0422
0.0355
0.0090
0.0042
0.0000
building
0.0448
0.9382
0.0133
0.0034
0.0003
0.0000
low_veg
0.0821
0.0210
0.7326
0.1639
0.0003
0.0000
tree
0.0133
0.0027
0.0577
0.9263
0.0001
0.0000
car
0.3257
0.0274
0.0019
0.0015
0.6436
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.876
0.956
0.722
0.927
0.897
--
Recall/Completeness
0.909
0.938
0.733
0.926
0.644
--
F1
0.892
0.947
0.727
0.927
0.749
--

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
0.9421
0.0274
0.0263
0.0021
0.0021
0.0000
building
0.0353
0.9545
0.0094
0.0007
0.0001
0.0000
low_veg
0.0697
0.0084
0.7795
0.1425
0.0000
0.0000
tree
0.0044
0.0004
0.0459
0.9493
0.0000
0.0000
car
0.2066
0.0217
0.0005
0.0000
0.7711
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.912
0.976
0.764
0.951
0.943
--
Recall/Completeness
0.942
0.955
0.779
0.949
0.771
--
F1
0.927
0.965
0.772
0.950
0.849
--

Overall accuracy 0.931

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
0.8855
0.0305
0.0709
0.0103
0.0027
0.0000
building
0.0433
0.9359
0.0162
0.0042
0.0005
0.0000
low_veg
0.0445
0.0131
0.8475
0.0943
0.0007
0.0000
tree
0.0193
0.0050
0.0663
0.9093
0.0001
0.0000
car
0.3074
0.0120
0.0191
0.0041
0.6575
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.907
0.941
0.798
0.905
0.908
--
Recall/Completeness
0.886
0.936
0.847
0.909
0.657
--
F1
0.896
0.938
0.822
0.907
0.763
--

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
0.9102
0.0251
0.0605
0.0029
0.0013
0.0000
building
0.0321
0.9574
0.0087
0.0013
0.0005
0.0000
low_veg
0.0289
0.0057
0.8809
0.0843
0.0003
0.0000
tree
0.0065
0.0024
0.0460
0.9451
0.0000
0.0000
car
0.1854
0.0129
0.0113
0.0030
0.7874
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.943
0.959
0.842
0.928
0.946
--
Recall/Completeness
0.910
0.957
0.881
0.945
0.787
--
F1
0.926
0.958
0.861
0.936
0.859
--

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
0.8413
0.0728
0.0647
0.0178
0.0034
0.0000
building
0.0359
0.9341
0.0194
0.0100
0.0007
0.0000
low_veg
0.0775
0.0297
0.7515
0.1411
0.0002
0.0000
tree
0.0122
0.0069
0.0951
0.8857
0.0000
0.0000
car
0.2437
0.0353
0.0179
0.0017
0.7014
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.923
0.866
0.731
0.843
0.773
--
Recall/Completeness
0.841
0.934
0.751
0.886
0.701
--
F1
0.880
0.899
0.741
0.864
0.735
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8735
0.0608
0.0561
0.0072
0.0023
0.0000
building
0.0304
0.9464
0.0163
0.0064
0.0005
0.0000
low_veg
0.0756
0.0158
0.7943
0.1142
0.0001
0.0000
tree
0.0049
0.0029
0.0728
0.9194
0.0000
0.0000
car
0.1437
0.0387
0.0178
0.0000
0.7998
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.939
0.895
0.770
0.886
0.802
--
Recall/Completeness
0.874
0.946
0.794
0.919
0.800
--
F1
0.905
0.920
0.782
0.902
0.801
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8846
0.0244
0.0567
0.0311
0.0032
0.0000
building
0.0375
0.9477
0.0093
0.0055
0.0000
0.0000
low_veg
0.0488
0.0220
0.7610
0.1678
0.0002
0.0002
tree
0.0102
0.0055
0.0910
0.8933
0.0000
0.0000
car
0.2418
0.0002
0.0000
0.0093
0.7486
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.827
0.939
0.835
0.859
0.877
--
Recall/Completeness
0.885
0.948
0.761
0.893
0.749
--
F1
0.855
0.943
0.796
0.876
0.808
--

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
0.9124
0.0220
0.0483
0.0156
0.0018
0.0000
building
0.0347
0.9607
0.0032
0.0014
0.0000
0.0000
low_veg
0.0458
0.0127
0.7975
0.1435
0.0002
0.0002
tree
0.0055
0.0018
0.0750
0.9177
0.0000
0.0000
car
0.0910
0.0000
0.0000
0.0011
0.9078
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.855
0.962
0.865
0.888
0.915
--
Recall/Completeness
0.912
0.961
0.797
0.918
0.908
--
F1
0.883
0.961
0.830
0.902
0.911
--

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
0.8370
0.0317
0.0949
0.0311
0.0053
0.0000
building
0.0124
0.9697
0.0127
0.0052
0.0000
0.0000
low_veg
0.0854
0.0451
0.7141
0.1540
0.0012
0.0000
tree
0.0248
0.0113
0.1109
0.8529
0.0001
0.0000
car
0.2036
0.0000
0.0002
0.0040
0.7922
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.854
0.883
0.748
0.858
0.577
--
Recall/Completeness
0.837
0.970
0.714
0.853
0.792
--
F1
0.846
0.924
0.731
0.856
0.668
--

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
0.8574
0.0263
0.0909
0.0207
0.0047
0.0000
building
0.0079
0.9849
0.0062
0.0010
0.0000
0.0000
low_veg
0.0806
0.0298
0.7451
0.1432
0.0013
0.0000
tree
0.0173
0.0049
0.0972
0.8805
0.0000
0.0000
car
0.0831
0.0000
0.0000
0.0000
0.9169
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.879
0.919
0.771
0.880
0.559
--
Recall/Completeness
0.857
0.985
0.745
0.881
0.917
--
F1
0.868
0.951
0.758
0.880
0.695
--

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
0.8205
0.0257
0.1011
0.0507
0.0021
0.0000
building
0.0393
0.9171
0.0329
0.0104
0.0004
0.0000
low_veg
0.0225
0.0255
0.8328
0.1186
0.0005
0.0000
tree
0.0219
0.0036
0.1072
0.8673
0.0000
0.0000
car
0.1781
0.0019
0.0337
0.0040
0.7823
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.852
0.927
0.802
0.876
0.760
--
Recall/Completeness
0.820
0.917
0.833
0.867
0.782
--
F1
0.836
0.922
0.817
0.872
0.771
--

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
0.8546
0.0213
0.0931
0.0302
0.0008
0.0000
building
0.0354
0.9406
0.0206
0.0030
0.0003
0.0000
low_veg
0.0191
0.0210
0.8562
0.1032
0.0005
0.0000
tree
0.0147
0.0010
0.0923
0.8920
0.0000
0.0000
car
0.1048
0.0000
0.0179
0.0000
0.8772
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.880
0.945
0.831
0.902
0.808
--
Recall/Completeness
0.855
0.941
0.856
0.892
0.877
--
F1
0.867
0.943
0.843
0.897
0.841
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8737
0.0263
0.0493
0.0476
0.0030
0.0000
building
0.0106
0.9687
0.0144
0.0063
0.0000
0.0000
low_veg
0.0600
0.0261
0.7248
0.1880
0.0010
0.0000
tree
0.0062
0.0013
0.0453
0.9472
0.0000
0.0000
car
0.1227
0.0214
0.0140
0.0130
0.8288
0.0000
clutter
0.5758
0.4219
0.0013
0.0010
0.0000
0.0000
Precision/Correctness
0.846
0.876
0.819
0.892
0.810
-nan
Recall/Completeness
0.874
0.969
0.725
0.947
0.829
0.000
F1
0.860
0.920
0.769
0.919
0.819
-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
0.9065
0.0192
0.0447
0.0281
0.0015
0.0000
building
0.0074
0.9803
0.0106
0.0017
0.0000
0.0000
low_veg
0.0532
0.0161
0.7684
0.1612
0.0011
0.0000
tree
0.0022
0.0002
0.0335
0.9641
0.0000
0.0000
car
0.0409
0.0250
0.0036
0.0059
0.9246
0.0000
clutter
0.5823
0.4174
0.0000
0.0003
0.0000
0.0000
Precision/Correctness
0.864
0.897
0.852
0.918
0.845
-nan
Recall/Completeness
0.906
0.980
0.768
0.964
0.925
0.000
F1
0.885
0.937
0.808
0.941
0.883
-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
0.8993
0.0248
0.0368
0.0336
0.0056
0.0000
building
0.0338
0.9504
0.0079
0.0074
0.0005
0.0000
low_veg
0.0879
0.0305
0.7244
0.1564
0.0008
0.0000
tree
0.0193
0.0033
0.0515
0.9259
0.0001
0.0000
car
0.2726
0.0167
0.0193
0.0145
0.6770
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.899
0.918
0.838
0.859
0.785
--
Recall/Completeness
0.899
0.950
0.724
0.926
0.677
--
F1
0.899
0.934
0.777
0.891
0.727
--

Overall accuracy 0.878

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
0.9301
0.0183
0.0313
0.0165
0.0038
0.0000
building
0.0281
0.9662
0.0023
0.0031
0.0004
0.0000
low_veg
0.0841
0.0179
0.7706
0.1267
0.0008
0.0000
tree
0.0073
0.0009
0.0365
0.9553
0.0000
0.0000
car
0.1523
0.0199
0.0154
0.0051
0.8073
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.924
0.947
0.877
0.899
0.818
--
Recall/Completeness
0.930
0.966
0.771
0.955
0.807
--
F1
0.927
0.956
0.820
0.926
0.813
--

Overall accuracy 0.912

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
0.8386
0.0308
0.0844
0.0398
0.0065
0.0000
building
0.0779
0.8653
0.0401
0.0167
0.0000
0.0000
low_veg
0.0242
0.0118
0.8848
0.0790
0.0002
0.0000
tree
0.0192
0.0063
0.1518
0.8226
0.0000
0.0000
car
0.2150
0.0003
0.0004
0.0035
0.7808
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.801
0.937
0.862
0.834
0.737
--
Recall/Completeness
0.839
0.865
0.885
0.823
0.781
--
F1
0.819
0.900
0.873
0.829
0.758
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8680
0.0213
0.0768
0.0286
0.0054
0.0000
building
0.0726
0.8787
0.0349
0.0137
0.0000
0.0000
low_veg
0.0198
0.0075
0.9037
0.0687
0.0002
0.0000
tree
0.0129
0.0029
0.1296
0.8547
0.0000
0.0000
car
0.0888
0.0004
0.0000
0.0004
0.9103
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.829
0.960
0.883
0.858
0.738
--
Recall/Completeness
0.868
0.879
0.904
0.855
0.910
--
F1
0.848
0.918
0.893
0.856
0.815
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9412
0.0276
0.0168
0.0054
0.0090
0.0000
building
0.0783
0.9030
0.0151
0.0033
0.0003
0.0000
low_veg
0.0879
0.0312
0.6915
0.1888
0.0005
0.0000
tree
0.0130
0.0043
0.0719
0.9107
0.0001
0.0000
car
0.3357
0.0033
0.0058
0.0051
0.6501
0.0000
clutter
0.5867
0.1888
0.1055
0.0074
0.1116
0.0000
Precision/Correctness
0.838
0.952
0.881
0.695
0.862
-nan
Recall/Completeness
0.941
0.903
0.692
0.911
0.650
0.000
F1
0.886
0.927
0.775
0.788
0.741
-nan

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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9662
0.0181
0.0102
0.0018
0.0036
0.0000
building
0.0661
0.9226
0.0101
0.0010
0.0002
0.0000
low_veg
0.0663
0.0175
0.7374
0.1785
0.0002
0.0000
tree
0.0050
0.0010
0.0552
0.9387
0.0000
0.0000
car
0.2328
0.0014
0.0034
0.0043
0.7581
0.0000
clutter
0.6191
0.1638
0.0877
0.0046
0.1248
0.0000
Precision/Correctness
0.875
0.972
0.917
0.725
0.925
-nan
Recall/Completeness
0.966
0.923
0.737
0.939
0.758
0.000
F1
0.918
0.947
0.818
0.818
0.833
-nan

Overall accuracy 0.897

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
0.9429
0.0178
0.0210
0.0136
0.0046
0.0000
building
0.0224
0.9582
0.0126
0.0068
0.0001
0.0000
low_veg
0.0493
0.0180
0.6488
0.2837
0.0003
0.0000
tree
0.0044
0.0015
0.0209
0.9731
0.0000
0.0001
car
0.3117
0.0026
0.0070
0.0120
0.6667
0.0000
clutter
0.5491
0.3633
0.0268
0.0066
0.0499
0.0043
Precision/Correctness
0.921
0.925
0.861
0.681
0.719
0.847
Recall/Completeness
0.943
0.958
0.649
0.973
0.667
0.004
F1
0.932
0.941
0.740
0.801
0.692
0.009

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9651
0.0112
0.0143
0.0058
0.0036
0.0000
building
0.0161
0.9731
0.0076
0.0032
0.0001
0.0000
low_veg
0.0289
0.0070
0.6932
0.2707
0.0002
0.0000
tree
0.0015
0.0001
0.0140
0.9844
0.0000
0.0000
car
0.2043
0.0011
0.0073
0.0091
0.7781
0.0000
clutter
0.5337
0.3850
0.0182
0.0029
0.0552
0.0051
Precision/Correctness
0.943
0.944
0.902
0.721
0.719
0.989
Recall/Completeness
0.965
0.973
0.693
0.984
0.778
0.005
F1
0.954
0.958
0.784
0.832
0.748
0.010

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9752
0.0059
0.0108
0.0039
0.0042
0.0002
building
0.0575
0.9273
0.0085
0.0026
0.0039
0.0002
low_veg
0.0405
0.0032
0.9067
0.0490
0.0004
0.0002
tree
0.0310
0.0022
0.0662
0.9001
0.0005
0.0000
car
0.2194
0.1822
0.0057
0.0038
0.5889
0.0000
clutter
0.5334
0.1905
0.0469
0.0088
0.0519
0.1685
Precision/Correctness
0.881
0.941
0.895
0.911
0.823
0.975
Recall/Completeness
0.975
0.927
0.907
0.900
0.589
0.168
F1
0.926
0.934
0.901
0.906
0.687
0.287

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
0.9845
0.0039
0.0087
0.0007
0.0021
0.0000
building
0.0437
0.9482
0.0042
0.0002
0.0035
0.0002
low_veg
0.0214
0.0011
0.9378
0.0396
0.0001
0.0000
tree
0.0083
0.0003
0.0504
0.9410
0.0001
0.0000
car
0.1189
0.2026
0.0016
0.0008
0.6761
0.0000
clutter
0.4854
0.2129
0.0494
0.0026
0.0551
0.1947
Precision/Correctness
0.917
0.952
0.923
0.942
0.846
0.988
Recall/Completeness
0.985
0.948
0.938
0.941
0.676
0.195
F1
0.949
0.950
0.930
0.941
0.752
0.325

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
0.9638
0.0249
0.0077
0.0026
0.0008
0.0002
building
0.0706
0.8484
0.0728
0.0057
0.0025
0.0000
low_veg
0.0583
0.0145
0.8833
0.0436
0.0003
0.0001
tree
0.0396
0.0217
0.1083
0.8304
0.0000
0.0000
car
0.3605
0.0022
0.0027
0.0018
0.6328
0.0000
clutter
0.6527
0.0526
0.0007
0.0013
0.0282
0.2645
Precision/Correctness
0.835
0.935
0.793
0.902
0.758
0.994
Recall/Completeness
0.964
0.848
0.883
0.830
0.633
0.265
F1
0.895
0.890
0.836
0.865
0.690
0.418

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_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9746
0.0215
0.0032
0.0005
0.0003
0.0000
building
0.0628
0.8600
0.0718
0.0030
0.0024
0.0000
low_veg
0.0304
0.0075
0.9222
0.0398
0.0001
0.0001
tree
0.0151
0.0202
0.0947
0.8700
0.0000
0.0000
car
0.2181
0.0008
0.0012
0.0002
0.7796
0.0000
clutter
0.6173
0.0529
0.0001
0.0001
0.0311
0.2986
Precision/Correctness
0.866
0.948
0.811
0.926
0.763
0.999
Recall/Completeness
0.975
0.860
0.922
0.870
0.780
0.299
F1
0.917
0.902
0.863
0.897
0.771
0.460

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8535
0.0671
0.0601
0.0175
0.0018
0.0000
building
0.0344
0.9443
0.0161
0.0051
0.0002
0.0000
low_veg
0.0541
0.0104
0.7055
0.2300
0.0001
0.0000
tree
0.0291
0.0052
0.0908
0.8748
0.0001
0.0000
car
0.3181
0.0069
0.0002
0.0070
0.6679
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.871
0.924
0.812
0.765
0.904
--
Recall/Completeness
0.853
0.944
0.705
0.875
0.668
--
F1
0.862
0.934
0.755
0.816
0.768
--

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
0.8907
0.0635
0.0407
0.0043
0.0008
0.0000
building
0.0264
0.9655
0.0072
0.0006
0.0002
0.0000
low_veg
0.0391
0.0030
0.7369
0.2210
0.0000
0.0000
tree
0.0081
0.0010
0.0670
0.9239
0.0000
0.0000
car
0.1938
0.0068
0.0001
0.0013
0.7979
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.916
0.942
0.868
0.792
0.949
--
Recall/Completeness
0.891
0.966
0.737
0.924
0.798
--
F1
0.903
0.954
0.797
0.853
0.867
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9211
0.0232
0.0416
0.0111
0.0026
0.0004
building
0.0389
0.9238
0.0304
0.0056
0.0014
0.0000
low_veg
0.0609
0.0297
0.8221
0.0871
0.0003
0.0000
tree
0.0357
0.0032
0.1070
0.8541
0.0000
0.0000
car
0.3687
0.0023
0.0527
0.0057
0.5706
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.868
0.939
0.851
0.853
0.831
--
Recall/Completeness
0.921
0.924
0.822
0.854
0.571
--
F1
0.894
0.931
0.836
0.853
0.677
--

Overall accuracy 0.878

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
0.9434
0.0206
0.0305
0.0038
0.0013
0.0004
building
0.0294
0.9478
0.0195
0.0022
0.0012
0.0000
low_veg
0.0408
0.0257
0.8630
0.0705
0.0000
0.0000
tree
0.0153
0.0007
0.0861
0.8979
0.0000
0.0000
car
0.2649
0.0002
0.0518
0.0016
0.6815
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.914
0.951
0.888
0.890
0.881
--
Recall/Completeness
0.943
0.948
0.863
0.898
0.682
--
F1
0.928
0.950
0.875
0.894
0.769
--

Overall accuracy 0.912

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image