Detailled semantic labeling (2D) result


Name K. Nogueira et al.
Affiliation Universidade Federal de Minas Gerais, Brazil
Abbreviation UFMG_4
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
0.039
0.018
0.003
0.001
building
0.055
0.919
0.016
0.009
0.000
0.000
low_veg
0.065
0.024
0.767
0.143
0.000
0.000
tree
0.023
0.006
0.092
0.878
0.000
0.000
car
0.288
0.054
0.009
0.006
0.641
0.001
clutter
0.505
0.173
0.023
0.009
0.027
0.263
Precision/Correctness
0.858
0.933
0.821
0.838
0.867
0.823
Recall/Completeness
0.909
0.919
0.767
0.878
0.641
0.263
F1
0.883
0.926
0.793
0.858
0.737
0.399

Overall accuracy 0.865

Overall statistics, reference set: no_boundary


Number of non processed tiles by this participant: 0
↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.933
0.024
0.031
0.009
0.002
0.001
building
0.046
0.937
0.010
0.006
0.000
0.000
low_veg
0.053
0.015
0.804
0.128
0.000
0.000
tree
0.012
0.003
0.077
0.908
0.000
0.000
car
0.182
0.063
0.006
0.003
0.744
0.002
clutter
0.482
0.179
0.022
0.003
0.028
0.286
Precision/Correctness
0.890
0.952
0.855
0.868
0.895
0.831
Recall/Completeness
0.933
0.937
0.804
0.908
0.744
0.286
F1
0.911
0.945
0.829
0.888
0.813
0.426

Overall accuracy 0.894
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9470
0.0101
0.0246
0.0132
0.0036
0.0015
building
0.0369
0.9374
0.0216
0.0040
0.0000
0.0000
low_veg
0.0788
0.0214
0.7300
0.1691
0.0002
0.0005
tree
0.0309
0.0076
0.0970
0.8631
0.0001
0.0013
car
0.3467
0.0018
0.0025
0.0013
0.6477
0.0000
clutter
0.0005
0.0006
0.0004
0.0127
0.0000
0.9858
Precision/Correctness
0.853
0.957
0.836
0.823
0.911
0.958
Recall/Completeness
0.947
0.937
0.730
0.863
0.648
0.986
F1
0.898
0.947
0.779
0.843
0.757
0.972

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9688
0.0064
0.0164
0.0039
0.0029
0.0016
building
0.0287
0.9576
0.0122
0.0015
0.0000
0.0000
low_veg
0.0615
0.0127
0.7620
0.1631
0.0002
0.0005
tree
0.0162
0.0044
0.0837
0.8953
0.0000
0.0005
car
0.2028
0.0013
0.0018
0.0004
0.7937
0.0000
clutter
0.0005
0.0004
0.0002
0.0022
0.0000
0.9967
Precision/Correctness
0.895
0.975
0.868
0.847
0.924
0.969
Recall/Completeness
0.969
0.958
0.762
0.895
0.794
0.997
F1
0.931
0.966
0.811
0.870
0.854
0.983

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9285
0.0520
0.0100
0.0058
0.0037
0.0000
building
0.0589
0.9332
0.0054
0.0023
0.0002
0.0000
low_veg
0.1643
0.0512
0.4868
0.2960
0.0017
0.0000
tree
0.0505
0.0171
0.0769
0.8554
0.0001
0.0001
car
0.3676
0.0121
0.0022
0.0009
0.6172
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.849
0.966
0.692
0.717
0.873
--
Recall/Completeness
0.929
0.933
0.487
0.855
0.617
--
F1
0.887
0.949
0.572
0.780
0.723
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9527
0.0344
0.0075
0.0029
0.0025
0.0000
building
0.0449
0.9493
0.0043
0.0013
0.0002
0.0000
low_veg
0.1235
0.0208
0.5327
0.3212
0.0017
0.0000
tree
0.0209
0.0069
0.0746
0.8976
0.0000
0.0000
car
0.2623
0.0092
0.0009
0.0000
0.7276
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.885
0.981
0.722
0.745
0.901
--
Recall/Completeness
0.953
0.949
0.533
0.898
0.728
--
F1
0.918
0.965
0.613
0.814
0.805
--

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
0.9004
0.0532
0.0331
0.0106
0.0026
0.0002
building
0.0426
0.9353
0.0170
0.0039
0.0003
0.0009
low_veg
0.0896
0.0434
0.6973
0.1694
0.0002
0.0000
tree
0.0156
0.0043
0.0633
0.9168
0.0001
0.0000
car
0.3268
0.0218
0.0022
0.0020
0.6471
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.873
0.940
0.697
0.923
0.930
--
Recall/Completeness
0.900
0.935
0.697
0.917
0.647
--
F1
0.886
0.938
0.697
0.920
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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9302
0.0399
0.0255
0.0032
0.0011
0.0002
building
0.0323
0.9525
0.0128
0.0013
0.0002
0.0009
low_veg
0.0715
0.0321
0.7450
0.1514
0.0000
0.0000
tree
0.0055
0.0012
0.0513
0.9420
0.0000
0.0000
car
0.2146
0.0176
0.0016
0.0002
0.7659
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.912
0.960
0.736
0.946
0.965
--
Recall/Completeness
0.930
0.953
0.745
0.942
0.766
--
F1
0.921
0.956
0.740
0.944
0.854
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8917
0.0350
0.0598
0.0104
0.0019
0.0012
building
0.0259
0.9531
0.0164
0.0045
0.0001
0.0000
low_veg
0.0561
0.0205
0.8720
0.0511
0.0001
0.0001
tree
0.0245
0.0072
0.0900
0.8779
0.0001
0.0004
car
0.3387
0.0181
0.0185
0.0055
0.6192
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.907
0.928
0.794
0.936
0.941
--
Recall/Completeness
0.892
0.953
0.872
0.878
0.619
--
F1
0.899
0.940
0.831
0.906
0.747
--

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
0.9157
0.0275
0.0513
0.0037
0.0008
0.0010
building
0.0164
0.9726
0.0088
0.0021
0.0001
0.0000
low_veg
0.0394
0.0120
0.9087
0.0399
0.0000
0.0001
tree
0.0092
0.0032
0.0710
0.9163
0.0000
0.0002
car
0.2267
0.0198
0.0110
0.0031
0.7393
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.945
0.951
0.833
0.959
0.970
--
Recall/Completeness
0.916
0.973
0.909
0.916
0.739
--
F1
0.930
0.962
0.869
0.937
0.839
--

Overall accuracy 0.926

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8557
0.0645
0.0434
0.0228
0.0022
0.0113
building
0.0159
0.9575
0.0115
0.0147
0.0003
0.0000
low_veg
0.0946
0.0302
0.7163
0.1575
0.0014
0.0000
tree
0.0149
0.0096
0.0650
0.9104
0.0000
0.0000
car
0.2444
0.0476
0.0111
0.0036
0.6932
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.927
0.877
0.795
0.825
0.808
--
Recall/Completeness
0.856
0.957
0.716
0.910
0.693
--
F1
0.890
0.915
0.754
0.865
0.746
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8888
0.0512
0.0366
0.0100
0.0012
0.0122
building
0.0104
0.9712
0.0072
0.0109
0.0003
0.0000
low_veg
0.0910
0.0185
0.7627
0.1263
0.0016
0.0000
tree
0.0072
0.0053
0.0501
0.9374
0.0000
0.0000
car
0.1508
0.0510
0.0103
0.0000
0.7879
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.944
0.907
0.833
0.871
0.839
--
Recall/Completeness
0.889
0.971
0.763
0.937
0.788
--
F1
0.916
0.938
0.796
0.903
0.812
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8749
0.0395
0.0468
0.0368
0.0019
0.0000
building
0.1752
0.8040
0.0140
0.0066
0.0002
0.0000
low_veg
0.0537
0.0277
0.7174
0.2006
0.0005
0.0001
tree
0.0110
0.0074
0.1020
0.8795
0.0000
0.0000
car
0.2473
0.0000
0.0014
0.0169
0.7344
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.696
0.904
0.815
0.834
0.886
--
Recall/Completeness
0.875
0.804
0.717
0.879
0.734
--
F1
0.775
0.851
0.763
0.856
0.803
--

Overall accuracy 0.817

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
0.9018
0.0372
0.0387
0.0222
0.0001
0.0000
building
0.1832
0.8087
0.0061
0.0018
0.0002
0.0000
low_veg
0.0505
0.0201
0.7509
0.1780
0.0004
0.0001
tree
0.0055
0.0032
0.0871
0.9041
0.0000
0.0000
car
0.1302
0.0000
0.0007
0.0126
0.8566
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.711
0.928
0.845
0.862
0.946
--
Recall/Completeness
0.902
0.809
0.751
0.904
0.857
--
F1
0.795
0.864
0.795
0.882
0.899
--

Overall accuracy 0.841

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
0.8430
0.0388
0.0756
0.0339
0.0044
0.0043
building
0.0170
0.9619
0.0154
0.0055
0.0000
0.0000
low_veg
0.0984
0.0552
0.7050
0.1410
0.0002
0.0001
tree
0.0321
0.0125
0.1236
0.8317
0.0000
0.0000
car
0.1501
0.0000
0.0009
0.0057
0.8433
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.832
0.862
0.745
0.862
0.675
--
Recall/Completeness
0.843
0.962
0.705
0.832
0.843
--
F1
0.837
0.909
0.724
0.847
0.750
--

Overall accuracy 0.825

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
0.8624
0.0345
0.0713
0.0234
0.0039
0.0045
building
0.0114
0.9796
0.0071
0.0019
0.0000
0.0000
low_veg
0.0929
0.0411
0.7363
0.1294
0.0002
0.0000
tree
0.0240
0.0060
0.1122
0.8577
0.0000
0.0000
car
0.0425
0.0000
0.0000
0.0021
0.9554
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.858
0.894
0.766
0.884
0.665
--
Recall/Completeness
0.862
0.980
0.736
0.858
0.955
--
F1
0.860
0.935
0.751
0.871
0.784
--

Overall accuracy 0.850

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
0.8146
0.0338
0.0980
0.0502
0.0022
0.0012
building
0.0396
0.9265
0.0207
0.0128
0.0003
0.0001
low_veg
0.0349
0.0152
0.8168
0.1332
0.0000
0.0000
tree
0.0247
0.0055
0.1204
0.8494
0.0000
0.0000
car
0.2025
0.0020
0.0027
0.0022
0.7907
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.824
0.934
0.794
0.863
0.818
--
Recall/Completeness
0.815
0.927
0.817
0.849
0.791
--
F1
0.819
0.930
0.805
0.856
0.804
--

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_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8472
0.0306
0.0887
0.0313
0.0009
0.0012
building
0.0343
0.9497
0.0105
0.0051
0.0003
0.0001
low_veg
0.0309
0.0095
0.8409
0.1187
0.0000
0.0000
tree
0.0173
0.0022
0.1068
0.8737
0.0000
0.0000
car
0.1254
0.0000
0.0001
0.0000
0.8745
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.854
0.954
0.821
0.887
0.884
--
Recall/Completeness
0.847
0.950
0.841
0.874
0.874
--
F1
0.851
0.952
0.831
0.880
0.879
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8607
0.0275
0.0688
0.0402
0.0028
0.0000
building
0.0164
0.9446
0.0314
0.0076
0.0000
0.0000
low_veg
0.0780
0.0177
0.7526
0.1516
0.0000
0.0000
tree
0.0086
0.0014
0.0578
0.9321
0.0000
0.0000
car
0.1816
0.0090
0.0472
0.0255
0.7366
0.0000
clutter
0.9545
0.0289
0.0149
0.0017
0.0000
0.0000
Precision/Correctness
0.783
0.937
0.771
0.908
0.843
-nan
Recall/Completeness
0.861
0.945
0.753
0.932
0.737
0.000
F1
0.820
0.941
0.762
0.920
0.786
-nan

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8893
0.0215
0.0641
0.0234
0.0016
0.0000
building
0.0131
0.9582
0.0254
0.0033
0.0000
0.0000
low_veg
0.0710
0.0084
0.7973
0.1233
0.0000
0.0000
tree
0.0038
0.0003
0.0448
0.9511
0.0000
0.0000
car
0.0965
0.0070
0.0381
0.0169
0.8416
0.0000
clutter
0.9635
0.0256
0.0105
0.0005
0.0000
0.0000
Precision/Correctness
0.801
0.960
0.802
0.934
0.886
-nan
Recall/Completeness
0.889
0.958
0.797
0.951
0.842
0.000
F1
0.843
0.959
0.800
0.942
0.863
-nan

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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8918
0.0263
0.0476
0.0304
0.0040
0.0000
building
0.0306
0.9452
0.0170
0.0067
0.0004
0.0000
low_veg
0.1062
0.0368
0.7049
0.1520
0.0001
0.0000
tree
0.0217
0.0039
0.0659
0.9083
0.0001
0.0000
car
0.3301
0.0213
0.0174
0.0132
0.6180
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.886
0.908
0.792
0.863
0.827
--
Recall/Completeness
0.892
0.945
0.705
0.908
0.618
--
F1
0.889
0.926
0.746
0.885
0.707
--

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_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9228
0.0206
0.0398
0.0143
0.0025
0.0000
building
0.0239
0.9641
0.0089
0.0027
0.0004
0.0000
low_veg
0.1027
0.0260
0.7456
0.1256
0.0001
0.0000
tree
0.0093
0.0013
0.0501
0.9392
0.0000
0.0000
car
0.2313
0.0297
0.0111
0.0040
0.7239
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.912
0.934
0.834
0.901
0.866
--
Recall/Completeness
0.923
0.964
0.746
0.939
0.724
--
F1
0.917
0.949
0.787
0.920
0.789
--

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_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8243
0.0325
0.0951
0.0421
0.0056
0.0005
building
0.0815
0.8663
0.0297
0.0224
0.0001
0.0000
low_veg
0.0319
0.0209
0.8784
0.0684
0.0004
0.0000
tree
0.0216
0.0065
0.1447
0.8270
0.0001
0.0000
car
0.2155
0.0000
0.0129
0.0074
0.7642
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.775
0.916
0.865
0.845
0.741
--
Recall/Completeness
0.824
0.866
0.878
0.827
0.764
--
F1
0.799
0.890
0.872
0.836
0.753
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8540
0.0246
0.0861
0.0307
0.0041
0.0005
building
0.0763
0.8806
0.0233
0.0198
0.0001
0.0000
low_veg
0.0280
0.0165
0.8990
0.0560
0.0004
0.0000
tree
0.0150
0.0036
0.1247
0.8566
0.0001
0.0000
car
0.1223
0.0000
0.0068
0.0043
0.8665
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.801
0.937
0.887
0.871
0.748
--
Recall/Completeness
0.854
0.881
0.899
0.857
0.867
--
F1
0.827
0.908
0.893
0.864
0.803
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9392
0.0299
0.0178
0.0083
0.0047
0.0000
building
0.0729
0.9064
0.0148
0.0059
0.0001
0.0000
low_veg
0.0911
0.0251
0.6769
0.2068
0.0001
0.0001
tree
0.0136
0.0051
0.0388
0.9423
0.0001
0.0000
car
0.2958
0.0092
0.0085
0.0061
0.6805
0.0000
clutter
0.7164
0.1123
0.1015
0.0147
0.0471
0.0081
Precision/Correctness
0.843
0.953
0.897
0.675
0.930
0.547
Recall/Completeness
0.939
0.906
0.677
0.942
0.681
0.008
F1
0.888
0.929
0.771
0.787
0.786
0.016

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
0.9613
0.0227
0.0117
0.0032
0.0011
0.0000
building
0.0599
0.9285
0.0088
0.0028
0.0000
0.0000
low_veg
0.0718
0.0145
0.7199
0.1936
0.0000
0.0001
tree
0.0067
0.0020
0.0252
0.9661
0.0000
0.0000
car
0.1888
0.0088
0.0071
0.0044
0.7909
0.0000
clutter
0.7279
0.1142
0.0861
0.0078
0.0562
0.0077
Precision/Correctness
0.879
0.970
0.933
0.709
0.976
0.470
Recall/Completeness
0.961
0.929
0.720
0.966
0.791
0.008
F1
0.918
0.949
0.813
0.818
0.874
0.015

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
0.9422
0.0161
0.0210
0.0163
0.0044
0.0000
building
0.0805
0.9017
0.0094
0.0072
0.0011
0.0000
low_veg
0.0628
0.0312
0.5884
0.3174
0.0002
0.0000
tree
0.0051
0.0062
0.0156
0.9731
0.0000
0.0000
car
0.2878
0.0025
0.0041
0.0112
0.6921
0.0023
clutter
0.5708
0.3095
0.0193
0.0157
0.0561
0.0287
Precision/Correctness
0.884
0.920
0.863
0.654
0.707
0.951
Recall/Completeness
0.942
0.902
0.588
0.973
0.692
0.029
F1
0.912
0.911
0.700
0.782
0.700
0.056

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9632
0.0099
0.0153
0.0082
0.0033
0.0000
building
0.0767
0.9133
0.0055
0.0035
0.0010
0.0000
low_veg
0.0401
0.0175
0.6371
0.3052
0.0001
0.0000
tree
0.0013
0.0043
0.0109
0.9835
0.0000
0.0000
car
0.1887
0.0003
0.0050
0.0079
0.7954
0.0027
clutter
0.5597
0.3230
0.0184
0.0066
0.0597
0.0327
Precision/Correctness
0.904
0.942
0.898
0.692
0.709
0.960
Recall/Completeness
0.963
0.913
0.637
0.984
0.795
0.033
F1
0.933
0.927
0.746
0.812
0.749
0.063

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9738
0.0090
0.0091
0.0045
0.0031
0.0005
building
0.0630
0.9253
0.0056
0.0035
0.0013
0.0013
low_veg
0.0530
0.0055
0.8955
0.0458
0.0002
0.0000
tree
0.0428
0.0037
0.0552
0.8979
0.0004
0.0000
car
0.2039
0.2482
0.0079
0.0067
0.5294
0.0039
clutter
0.4501
0.3334
0.0558
0.0141
0.0169
0.1297
Precision/Correctness
0.878
0.911
0.908
0.909
0.897
0.858
Recall/Completeness
0.974
0.925
0.895
0.898
0.529
0.130
F1
0.923
0.918
0.902
0.903
0.666
0.225

Overall accuracy 0.896

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
0.9849
0.0057
0.0068
0.0011
0.0012
0.0004
building
0.0529
0.9414
0.0028
0.0008
0.0010
0.0012
low_veg
0.0311
0.0024
0.9306
0.0357
0.0002
0.0000
tree
0.0160
0.0007
0.0418
0.9414
0.0000
0.0000
car
0.1226
0.2826
0.0041
0.0021
0.5839
0.0048
clutter
0.3916
0.3810
0.0615
0.0059
0.0178
0.1422
Precision/Correctness
0.912
0.925
0.933
0.943
0.927
0.873
Recall/Completeness
0.985
0.941
0.931
0.941
0.584
0.142
F1
0.947
0.933
0.932
0.942
0.716
0.245

Overall accuracy 0.922

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
0.9842
0.0037
0.0066
0.0035
0.0013
0.0007
building
0.0890
0.8385
0.0123
0.0574
0.0021
0.0008
low_veg
0.0769
0.0083
0.8612
0.0517
0.0019
0.0000
tree
0.0461
0.0059
0.1375
0.8104
0.0000
0.0000
car
0.3130
0.0136
0.0076
0.0029
0.6548
0.0081
clutter
0.5362
0.1342
0.0110
0.0016
0.0449
0.2721
Precision/Correctness
0.834
0.958
0.862
0.765
0.688
0.961
Recall/Completeness
0.984
0.838
0.861
0.810
0.655
0.272
F1
0.903
0.894
0.862
0.787
0.671
0.424

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
0.9935
0.0019
0.0029
0.0010
0.0006
0.0001
building
0.0783
0.8516
0.0099
0.0577
0.0018
0.0007
low_veg
0.0439
0.0046
0.9000
0.0496
0.0018
0.0000
tree
0.0216
0.0022
0.1273
0.8489
0.0000
0.0000
car
0.1983
0.0178
0.0048
0.0012
0.7677
0.0103
clutter
0.5022
0.1434
0.0108
0.0013
0.0480
0.2944
Precision/Correctness
0.865
0.966
0.885
0.774
0.688
0.975
Recall/Completeness
0.993
0.852
0.900
0.849
0.768
0.294
F1
0.925
0.905
0.892
0.810
0.726
0.452

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8853
0.0311
0.0638
0.0190
0.0006
0.0001
building
0.0446
0.9321
0.0184
0.0047
0.0003
0.0000
low_veg
0.0566
0.0155
0.7074
0.2205
0.0001
0.0000
tree
0.0346
0.0066
0.1146
0.8441
0.0000
0.0001
car
0.3232
0.0346
0.0020
0.0120
0.6282
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.859
0.949
0.786
0.765
0.947
--
Recall/Completeness
0.885
0.932
0.707
0.844
0.628
--
F1
0.872
0.940
0.745
0.802
0.755
--

Overall accuracy 0.842

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
0.9228
0.0260
0.0456
0.0055
0.0002
0.0000
building
0.0311
0.9602
0.0075
0.0009
0.0003
0.0000
low_veg
0.0401
0.0088
0.7372
0.2139
0.0000
0.0000
tree
0.0129
0.0019
0.0910
0.8941
0.0000
0.0001
car
0.1947
0.0427
0.0000
0.0040
0.7586
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.909
0.967
0.841
0.791
0.973
--
Recall/Completeness
0.923
0.960
0.737
0.894
0.759
--
F1
0.916
0.964
0.786
0.839
0.852
--

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
0.9078
0.0320
0.0439
0.0140
0.0016
0.0007
building
0.0361
0.9249
0.0300
0.0084
0.0005
0.0000
low_veg
0.0466
0.0144
0.8397
0.0991
0.0001
0.0000
tree
0.0273
0.0054
0.1073
0.8599
0.0001
0.0000
car
0.2740
0.0531
0.0227
0.0079
0.6403
0.0020
clutter
--
--
--
--
--
--
Precision/Correctness
0.890
0.943
0.853
0.833
0.910
--
Recall/Completeness
0.908
0.925
0.840
0.860
0.640
--
F1
0.899
0.934
0.846
0.846
0.752
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9323
0.0300
0.0311
0.0053
0.0005
0.0007
building
0.0277
0.9502
0.0173
0.0044
0.0004
0.0000
low_veg
0.0304
0.0088
0.8788
0.0820
0.0001
0.0000
tree
0.0117
0.0027
0.0868
0.8987
0.0001
0.0000
car
0.1729
0.0645
0.0125
0.0043
0.7430
0.0028
clutter
--
--
--
--
--
--
Precision/Correctness
0.929
0.956
0.892
0.871
0.950
--
Recall/Completeness
0.932
0.950
0.879
0.899
0.743
--
F1
0.930
0.953
0.885
0.885
0.834
--

Overall accuracy 0.915

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image