Detailled semantic labeling (2D) result


Name H. Wang
Affiliation NLPR, Inst. of Automation, China
Abbreviation GSN2
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
0.911
0.030
0.040
0.016
0.004
0.000
building
0.047
0.923
0.024
0.006
0.000
0.000
low_veg
0.053
0.019
0.788
0.140
0.000
0.000
tree
0.019
0.004
0.090
0.887
0.000
0.000
car
0.252
0.050
0.008
0.004
0.686
0.000
clutter
0.249
0.699
0.027
0.006
0.019
0.000
Precision/Correctness
0.883
0.923
0.819
0.847
0.847
-nan
Recall/Completeness
0.911
0.923
0.788
0.887
0.686
0.000
F1
0.897
0.923
0.803
0.867
0.758
-nan

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
0.935
0.024
0.032
0.007
0.002
0.000
building
0.038
0.944
0.016
0.002
0.000
0.000
low_veg
0.041
0.012
0.822
0.125
0.000
0.000
tree
0.009
0.002
0.073
0.917
0.000
0.000
car
0.147
0.057
0.005
0.002
0.789
0.000
clutter
0.207
0.748
0.024
0.002
0.019
0.000
Precision/Correctness
0.916
0.939
0.854
0.877
0.888
-nan
Recall/Completeness
0.935
0.944
0.822
0.917
0.789
0.000
F1
0.925
0.941
0.837
0.896
0.836
-nan

Overall accuracy 0.900
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9517
0.0139
0.0183
0.0108
0.0051
0.0000
building
0.0347
0.9394
0.0224
0.0035
0.0000
0.0000
low_veg
0.0789
0.0161
0.7310
0.1738
0.0002
0.0000
tree
0.0284
0.0052
0.0887
0.8776
0.0002
0.0000
car
0.2793
0.0189
0.0003
0.0014
0.7001
0.0000
clutter
0.0094
0.9338
0.0452
0.0116
0.0000
0.0000
Precision/Correctness
0.860
0.891
0.846
0.824
0.887
-nan
Recall/Completeness
0.952
0.939
0.731
0.878
0.700
0.000
F1
0.904
0.915
0.784
0.850
0.783
-nan

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9729
0.0111
0.0096
0.0026
0.0037
0.0000
building
0.0256
0.9611
0.0128
0.0005
0.0000
0.0000
low_veg
0.0612
0.0102
0.7614
0.1671
0.0002
0.0000
tree
0.0133
0.0020
0.0751
0.9096
0.0001
0.0000
car
0.1435
0.0204
0.0000
0.0004
0.8357
0.0000
clutter
0.0096
0.9414
0.0466
0.0024
0.0000
0.0000
Precision/Correctness
0.902
0.903
0.879
0.847
0.911
-nan
Recall/Completeness
0.973
0.961
0.761
0.910
0.836
0.000
F1
0.936
0.931
0.816
0.877
0.872
-nan

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9473
0.0314
0.0142
0.0045
0.0026
0.0000
building
0.0795
0.9089
0.0092
0.0015
0.0008
0.0000
low_veg
0.1017
0.0507
0.5787
0.2672
0.0017
0.0000
tree
0.0446
0.0103
0.0666
0.8781
0.0003
0.0000
car
0.3494
0.0096
0.0038
0.0019
0.6353
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.831
0.977
0.672
0.749
0.875
--
Recall/Completeness
0.947
0.909
0.579
0.878
0.635
--
F1
0.885
0.942
0.622
0.808
0.736
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9668
0.0206
0.0101
0.0014
0.0011
0.0000
building
0.0625
0.9290
0.0072
0.0005
0.0009
0.0000
low_veg
0.0651
0.0341
0.6093
0.2903
0.0012
0.0000
tree
0.0178
0.0044
0.0553
0.9224
0.0000
0.0000
car
0.2249
0.0078
0.0025
0.0007
0.7642
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.866
0.986
0.715
0.779
0.907
--
Recall/Completeness
0.967
0.929
0.609
0.922
0.764
--
F1
0.914
0.957
0.658
0.844
0.830
--

Overall accuracy 0.925

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
0.9012
0.0493
0.0359
0.0083
0.0053
0.0000
building
0.0437
0.9335
0.0195
0.0031
0.0002
0.0000
low_veg
0.0712
0.0163
0.7628
0.1493
0.0004
0.0000
tree
0.0138
0.0021
0.0680
0.9160
0.0001
0.0000
car
0.2798
0.0227
0.0022
0.0008
0.6944
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.882
0.953
0.698
0.933
0.885
--
Recall/Completeness
0.901
0.934
0.763
0.916
0.694
--
F1
0.892
0.943
0.729
0.924
0.778
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9332
0.0374
0.0248
0.0026
0.0021
0.0000
building
0.0322
0.9548
0.0125
0.0005
0.0000
0.0000
low_veg
0.0560
0.0074
0.8015
0.1349
0.0001
0.0000
tree
0.0046
0.0003
0.0536
0.9415
0.0000
0.0000
car
0.1808
0.0232
0.0006
0.0000
0.7955
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.920
0.969
0.747
0.953
0.948
--
Recall/Completeness
0.933
0.955
0.802
0.941
0.795
--
F1
0.927
0.962
0.773
0.947
0.865
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8957
0.0259
0.0669
0.0082
0.0033
0.0000
building
0.0507
0.9246
0.0202
0.0045
0.0000
0.0000
low_veg
0.0459
0.0104
0.8605
0.0822
0.0010
0.0000
tree
0.0248
0.0042
0.0664
0.9045
0.0001
0.0000
car
0.2729
0.0146
0.0113
0.0035
0.6977
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.900
0.949
0.802
0.916
0.904
--
Recall/Completeness
0.896
0.925
0.860
0.904
0.698
--
F1
0.898
0.937
0.830
0.910
0.788
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9175
0.0223
0.0568
0.0017
0.0016
0.0000
building
0.0384
0.9492
0.0109
0.0015
0.0000
0.0000
low_veg
0.0297
0.0048
0.8917
0.0734
0.0004
0.0000
tree
0.0098
0.0020
0.0471
0.9411
0.0000
0.0000
car
0.1611
0.0160
0.0070
0.0025
0.8134
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.937
0.964
0.845
0.937
0.946
--
Recall/Completeness
0.918
0.949
0.892
0.941
0.813
--
F1
0.927
0.956
0.868
0.939
0.875
--

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
0.8565
0.0717
0.0493
0.0183
0.0041
0.0000
building
0.0337
0.9355
0.0202
0.0101
0.0005
0.0000
low_veg
0.0893
0.0324
0.7381
0.1399
0.0003
0.0000
tree
0.0136
0.0073
0.0928
0.8863
0.0001
0.0000
car
0.1703
0.0426
0.0043
0.0024
0.7803
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.921
0.865
0.754
0.843
0.760
--
Recall/Completeness
0.857
0.935
0.738
0.886
0.780
--
F1
0.887
0.899
0.746
0.864
0.770
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8864
0.0631
0.0403
0.0075
0.0027
0.0000
building
0.0275
0.9493
0.0160
0.0069
0.0004
0.0000
low_veg
0.0867
0.0236
0.7743
0.1151
0.0003
0.0000
tree
0.0064
0.0032
0.0711
0.9192
0.0000
0.0000
car
0.0814
0.0473
0.0019
0.0000
0.8694
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.937
0.887
0.796
0.885
0.795
--
Recall/Completeness
0.886
0.949
0.774
0.919
0.869
--
F1
0.911
0.917
0.785
0.902
0.831
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8809
0.0315
0.0529
0.0314
0.0033
0.0000
building
0.0241
0.9532
0.0158
0.0069
0.0000
0.0000
low_veg
0.0351
0.0190
0.7668
0.1790
0.0001
0.0000
tree
0.0085
0.0048
0.0921
0.8946
0.0000
0.0000
car
0.1610
0.0008
0.0000
0.0026
0.8355
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.870
0.939
0.833
0.852
0.892
--
Recall/Completeness
0.881
0.953
0.767
0.895
0.836
--
F1
0.875
0.946
0.798
0.873
0.863
--

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_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9092
0.0295
0.0442
0.0158
0.0013
0.0000
building
0.0224
0.9685
0.0069
0.0022
0.0000
0.0000
low_veg
0.0321
0.0127
0.8006
0.1547
0.0000
0.0000
tree
0.0042
0.0015
0.0763
0.9180
0.0000
0.0000
car
0.0411
0.0000
0.0000
0.0000
0.9589
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.895
0.957
0.863
0.880
0.949
--
Recall/Completeness
0.909
0.968
0.801
0.918
0.959
--
F1
0.902
0.963
0.831
0.899
0.954
--

Overall accuracy 0.893

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
0.8451
0.0275
0.0895
0.0316
0.0064
0.0000
building
0.0188
0.9513
0.0238
0.0062
0.0000
0.0000
low_veg
0.0757
0.0370
0.7356
0.1513
0.0003
0.0000
tree
0.0217
0.0075
0.1386
0.8321
0.0000
0.0000
car
0.1315
0.0000
0.0000
0.0038
0.8647
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.865
0.902
0.725
0.856
0.596
--
Recall/Completeness
0.845
0.951
0.736
0.832
0.865
--
F1
0.855
0.926
0.730
0.844
0.706
--

Overall accuracy 0.832

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
0.8645
0.0241
0.0845
0.0214
0.0055
0.0000
building
0.0124
0.9737
0.0121
0.0019
0.0000
0.0000
low_veg
0.0702
0.0253
0.7622
0.1420
0.0003
0.0000
tree
0.0139
0.0030
0.1238
0.8593
0.0000
0.0000
car
0.0364
0.0000
0.0000
0.0000
0.9636
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.892
0.929
0.749
0.877
0.580
--
Recall/Completeness
0.865
0.974
0.762
0.859
0.964
--
F1
0.878
0.951
0.755
0.868
0.724
--

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_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8227
0.0204
0.1014
0.0530
0.0025
0.0000
building
0.0209
0.9374
0.0287
0.0128
0.0001
0.0000
low_veg
0.0200
0.0127
0.8445
0.1227
0.0000
0.0000
tree
0.0220
0.0038
0.1094
0.8648
0.0000
0.0000
car
0.1639
0.0001
0.0027
0.0032
0.8301
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.875
0.952
0.804
0.871
0.817
--
Recall/Completeness
0.823
0.937
0.844
0.865
0.830
--
F1
0.848
0.945
0.824
0.868
0.823
--

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
0.8555
0.0170
0.0935
0.0331
0.0009
0.0000
building
0.0172
0.9633
0.0153
0.0042
0.0000
0.0000
low_veg
0.0164
0.0079
0.8680
0.1077
0.0000
0.0000
tree
0.0149
0.0014
0.0950
0.8887
0.0000
0.0000
car
0.0993
0.0000
0.0001
0.0001
0.9004
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.905
0.970
0.833
0.896
0.910
--
Recall/Completeness
0.855
0.963
0.868
0.889
0.900
--
F1
0.879
0.966
0.850
0.893
0.905
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8620
0.0437
0.0487
0.0422
0.0034
0.0000
building
0.0102
0.9675
0.0155
0.0068
0.0000
0.0000
low_veg
0.0440
0.0297
0.7497
0.1766
0.0000
0.0000
tree
0.0070
0.0013
0.0467
0.9450
0.0000
0.0000
car
0.0762
0.0007
0.0296
0.0114
0.8821
0.0000
clutter
0.0078
0.9909
0.0011
0.0002
0.0000
0.0000
Precision/Correctness
0.928
0.791
0.821
0.898
0.841
-nan
Recall/Completeness
0.862
0.968
0.750
0.945
0.882
0.000
F1
0.894
0.870
0.784
0.921
0.861
-nan

Overall accuracy 0.868

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
0.8953
0.0363
0.0428
0.0243
0.0013
0.0000
building
0.0072
0.9825
0.0087
0.0016
0.0000
0.0000
low_veg
0.0383
0.0210
0.7923
0.1484
0.0000
0.0000
tree
0.0028
0.0003
0.0341
0.9628
0.0000
0.0000
car
0.0152
0.0000
0.0194
0.0043
0.9611
0.0000
clutter
0.0075
0.9925
0.0000
0.0000
0.0000
0.0000
Precision/Correctness
0.949
0.806
0.857
0.925
0.914
-nan
Recall/Completeness
0.895
0.983
0.792
0.963
0.961
0.000
F1
0.921
0.886
0.824
0.944
0.937
-nan

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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8996
0.0190
0.0438
0.0310
0.0066
0.0000
building
0.0343
0.9445
0.0134
0.0078
0.0001
0.0000
low_veg
0.0748
0.0299
0.7563
0.1388
0.0002
0.0000
tree
0.0199
0.0029
0.0572
0.9198
0.0002
0.0000
car
0.2228
0.0078
0.0331
0.0091
0.7272
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.907
0.928
0.821
0.871
0.783
--
Recall/Completeness
0.900
0.945
0.756
0.920
0.727
--
F1
0.903
0.936
0.787
0.895
0.754
--

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_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9305
0.0138
0.0365
0.0149
0.0043
0.0000
building
0.0279
0.9645
0.0044
0.0031
0.0000
0.0000
low_veg
0.0705
0.0209
0.7981
0.1104
0.0002
0.0000
tree
0.0079
0.0007
0.0417
0.9497
0.0001
0.0000
car
0.1304
0.0103
0.0278
0.0017
0.8298
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.932
0.952
0.863
0.910
0.824
--
Recall/Completeness
0.930
0.965
0.798
0.950
0.830
--
F1
0.931
0.958
0.829
0.929
0.827
--

Overall accuracy 0.915

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
0.8199
0.0289
0.1122
0.0327
0.0063
0.0000
building
0.0469
0.8687
0.0714
0.0131
0.0000
0.0000
low_veg
0.0259
0.0111
0.8737
0.0893
0.0002
0.0000
tree
0.0171
0.0055
0.1401
0.8372
0.0000
0.0000
car
0.1503
0.0000
0.0012
0.0016
0.8469
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.830
0.942
0.846
0.829
0.763
--
Recall/Completeness
0.820
0.869
0.874
0.837
0.847
--
F1
0.825
0.904
0.860
0.833
0.803
--

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
0.8506
0.0191
0.1039
0.0220
0.0043
0.0000
building
0.0433
0.8827
0.0651
0.0090
0.0000
0.0000
low_veg
0.0219
0.0073
0.8934
0.0772
0.0002
0.0000
tree
0.0103
0.0023
0.1205
0.8669
0.0000
0.0000
car
0.0544
0.0000
0.0002
0.0000
0.9454
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.858
0.964
0.867
0.854
0.788
--
Recall/Completeness
0.851
0.883
0.893
0.867
0.945
--
F1
0.854
0.921
0.880
0.860
0.859
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9439
0.0237
0.0206
0.0057
0.0061
0.0000
building
0.0840
0.8927
0.0195
0.0035
0.0002
0.0000
low_veg
0.0849
0.0212
0.7132
0.1806
0.0002
0.0000
tree
0.0118
0.0046
0.0670
0.9165
0.0001
0.0000
car
0.3258
0.0070
0.0088
0.0056
0.6529
0.0000
clutter
0.7307
0.1165
0.1146
0.0092
0.0290
0.0000
Precision/Correctness
0.835
0.961
0.871
0.704
0.908
-nan
Recall/Completeness
0.944
0.893
0.713
0.917
0.653
0.000
F1
0.886
0.926
0.784
0.796
0.760
-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
0.9658
0.0171
0.0130
0.0020
0.0021
0.0000
building
0.0700
0.9156
0.0134
0.0009
0.0001
0.0000
low_veg
0.0656
0.0107
0.7541
0.1695
0.0000
0.0000
tree
0.0041
0.0016
0.0492
0.9451
0.0000
0.0000
car
0.2106
0.0064
0.0064
0.0042
0.7725
0.0000
clutter
0.7654
0.1027
0.0912
0.0067
0.0340
0.0000
Precision/Correctness
0.872
0.977
0.910
0.736
0.960
-nan
Recall/Completeness
0.966
0.916
0.754
0.945
0.772
0.000
F1
0.917
0.945
0.825
0.827
0.856
-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
0.9444
0.0166
0.0211
0.0125
0.0054
0.0000
building
0.0226
0.9566
0.0141
0.0066
0.0001
0.0000
low_veg
0.0675
0.0147
0.6493
0.2682
0.0005
0.0000
tree
0.0045
0.0011
0.0320
0.9625
0.0000
0.0000
car
0.2307
0.0555
0.0057
0.0094
0.6987
0.0000
clutter
0.3012
0.6357
0.0333
0.0101
0.0198
0.0000
Precision/Correctness
0.933
0.900
0.847
0.691
0.759
-nan
Recall/Completeness
0.944
0.957
0.649
0.962
0.699
0.000
F1
0.939
0.927
0.735
0.804
0.728
-nan

Overall accuracy 0.874

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
0.9649
0.0112
0.0149
0.0051
0.0039
0.0000
building
0.0146
0.9741
0.0084
0.0029
0.0001
0.0000
low_veg
0.0494
0.0062
0.6901
0.2541
0.0002
0.0000
tree
0.0015
0.0001
0.0232
0.9752
0.0000
0.0000
car
0.1335
0.0618
0.0052
0.0064
0.7931
0.0000
clutter
0.2635
0.6877
0.0236
0.0050
0.0202
0.0000
Precision/Correctness
0.955
0.917
0.888
0.732
0.780
-nan
Recall/Completeness
0.965
0.974
0.690
0.975
0.793
0.000
F1
0.960
0.945
0.777
0.836
0.786
-nan

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9746
0.0094
0.0077
0.0033
0.0051
0.0000
building
0.0557
0.9315
0.0077
0.0049
0.0002
0.0000
low_veg
0.0395
0.0042
0.8921
0.0638
0.0005
0.0000
tree
0.0360
0.0023
0.0490
0.9119
0.0008
0.0000
car
0.1582
0.2377
0.0060
0.0042
0.5939
0.0000
clutter
0.3812
0.5380
0.0495
0.0101
0.0212
0.0000
Precision/Correctness
0.896
0.892
0.914
0.891
0.880
-nan
Recall/Completeness
0.975
0.931
0.892
0.912
0.594
0.000
F1
0.934
0.911
0.903
0.902
0.709
-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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9859
0.0060
0.0053
0.0007
0.0021
0.0000
building
0.0411
0.9540
0.0041
0.0006
0.0002
0.0000
low_veg
0.0220
0.0019
0.9199
0.0560
0.0002
0.0000
tree
0.0114
0.0003
0.0353
0.9529
0.0001
0.0000
car
0.0774
0.2637
0.0013
0.0013
0.6564
0.0000
clutter
0.3227
0.6037
0.0499
0.0026
0.0211
0.0000
Precision/Correctness
0.931
0.907
0.941
0.921
0.924
-nan
Recall/Completeness
0.986
0.954
0.920
0.953
0.656
0.000
F1
0.958
0.930
0.930
0.937
0.768
-nan

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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9730
0.0155
0.0064
0.0030
0.0021
0.0000
building
0.0827
0.8523
0.0600
0.0035
0.0014
0.0000
low_veg
0.0536
0.0155
0.8996
0.0309
0.0003
0.0000
tree
0.0383
0.0101
0.1243
0.8273
0.0000
0.0000
car
0.2418
0.0004
0.0016
0.0027
0.7535
0.0000
clutter
0.3406
0.6187
0.0024
0.0010
0.0372
0.0000
Precision/Correctness
0.866
0.852
0.807
0.925
0.747
-nan
Recall/Completeness
0.973
0.852
0.900
0.827
0.753
0.000
F1
0.917
0.852
0.851
0.874
0.750
-nan

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_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9847
0.0115
0.0020
0.0008
0.0010
0.0000
building
0.0751
0.8641
0.0580
0.0014
0.0014
0.0000
low_veg
0.0273
0.0094
0.9354
0.0279
0.0001
0.0000
tree
0.0152
0.0073
0.1125
0.8650
0.0000
0.0000
car
0.1208
0.0000
0.0006
0.0005
0.8780
0.0000
clutter
0.2956
0.6646
0.0017
0.0000
0.0380
0.0000
Precision/Correctness
0.895
0.867
0.826
0.947
0.762
-nan
Recall/Completeness
0.985
0.864
0.935
0.865
0.878
0.000
F1
0.938
0.866
0.877
0.904
0.816
-nan

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8615
0.0604
0.0620
0.0147
0.0014
0.0000
building
0.0332
0.9411
0.0203
0.0052
0.0002
0.0000
low_veg
0.0539
0.0081
0.7267
0.2112
0.0000
0.0000
tree
0.0284
0.0052
0.1070
0.8593
0.0001
0.0000
car
0.2913
0.0119
0.0024
0.0055
0.6888
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.874
0.931
0.797
0.778
0.924
--
Recall/Completeness
0.861
0.941
0.727
0.859
0.689
--
F1
0.868
0.936
0.760
0.816
0.789
--

Overall accuracy 0.848

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
0.8949
0.0584
0.0432
0.0029
0.0005
0.0000
building
0.0230
0.9673
0.0088
0.0007
0.0002
0.0000
low_veg
0.0369
0.0031
0.7545
0.2055
0.0000
0.0000
tree
0.0072
0.0015
0.0818
0.9095
0.0000
0.0000
car
0.1765
0.0147
0.0017
0.0007
0.8064
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.924
0.946
0.853
0.802
0.966
--
Recall/Completeness
0.895
0.967
0.755
0.910
0.806
--
F1
0.909
0.957
0.801
0.852
0.879
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9131
0.0301
0.0401
0.0127
0.0040
0.0000
building
0.0350
0.9174
0.0390
0.0070
0.0017
0.0000
low_veg
0.0369
0.0190
0.8452
0.0986
0.0003
0.0000
tree
0.0200
0.0024
0.1077
0.8698
0.0001
0.0000
car
0.2768
0.0131
0.0196
0.0057
0.6849
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.905
0.943
0.850
0.838
0.805
--
Recall/Completeness
0.913
0.917
0.845
0.870
0.685
--
F1
0.909
0.930
0.848
0.854
0.740
--

Overall accuracy 0.885

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9371
0.0287
0.0286
0.0037
0.0019
0.0000
building
0.0265
0.9466
0.0237
0.0019
0.0014
0.0000
low_veg
0.0192
0.0167
0.8831
0.0810
0.0000
0.0000
tree
0.0064
0.0004
0.0863
0.9068
0.0000
0.0000
car
0.1850
0.0132
0.0114
0.0019
0.7885
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.944
0.952
0.889
0.878
0.868
--
Recall/Completeness
0.937
0.947
0.883
0.907
0.789
--
F1
0.941
0.949
0.886
0.892
0.826
--

Overall accuracy 0.918

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image