Detailled semantic labeling (2D) result


Name Paisitkriangkrai, Sherrah, Janney, van den Hengel (CNN and RF and CRF)
Affiliation Uni of Adelaide & DSTO,Australia
Abbreviation ADL_3
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
0.873
0.037
0.045
0.020
0.025
0.000
building
0.064
0.905
0.018
0.009
0.003
0.000
low_veg
0.060
0.027
0.769
0.141
0.002
0.000
tree
0.021
0.008
0.101
0.870
0.001
0.000
car
0.171
0.024
0.016
0.006
0.783
0.000
clutter
0.466
0.279
0.041
0.009
0.205
0.000
Precision/Correctness
0.856
0.919
0.803
0.836
0.500
-nan
Recall/Completeness
0.873
0.905
0.769
0.870
0.783
0.000
F1
0.864
0.912
0.786
0.853
0.610
-nan

Overall accuracy 0.849

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
0.907
0.029
0.036
0.011
0.017
0.000
building
0.057
0.922
0.012
0.007
0.002
0.000
low_veg
0.050
0.017
0.806
0.126
0.001
0.000
tree
0.012
0.004
0.085
0.900
0.000
0.000
car
0.131
0.026
0.014
0.003
0.827
0.000
clutter
0.450
0.298
0.036
0.004
0.212
0.000
Precision/Correctness
0.882
0.942
0.840
0.865
0.513
-nan
Recall/Completeness
0.907
0.922
0.806
0.900
0.827
0.000
F1
0.895
0.932
0.823
0.882
0.633
-nan

Overall accuracy 0.880
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8922
0.0123
0.0504
0.0147
0.0304
0.0000
building
0.0320
0.9397
0.0230
0.0038
0.0015
0.0000
low_veg
0.0708
0.0227
0.8044
0.0996
0.0025
0.0000
tree
0.0218
0.0081
0.1262
0.8433
0.0005
0.0000
car
0.2074
0.0014
0.0096
0.0031
0.7785
0.0000
clutter
0.9518
0.0205
0.0052
0.0225
0.0000
0.0000
Precision/Correctness
0.814
0.951
0.800
0.876
0.583
-nan
Recall/Completeness
0.892
0.940
0.804
0.843
0.778
0.000
F1
0.851
0.945
0.802
0.860
0.667
-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
0.9301
0.0080
0.0334
0.0061
0.0224
0.0000
building
0.0248
0.9586
0.0135
0.0019
0.0012
0.0000
low_veg
0.0595
0.0122
0.8347
0.0921
0.0016
0.0000
tree
0.0106
0.0044
0.1096
0.8752
0.0001
0.0000
car
0.1556
0.0003
0.0095
0.0003
0.8343
0.0000
clutter
0.9714
0.0180
0.0028
0.0078
0.0000
0.0000
Precision/Correctness
0.840
0.973
0.837
0.900
0.616
-nan
Recall/Completeness
0.930
0.959
0.835
0.875
0.834
0.000
F1
0.883
0.966
0.836
0.888
0.709
-nan

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8525
0.0997
0.0134
0.0060
0.0284
0.0000
building
0.0454
0.9433
0.0074
0.0020
0.0020
0.0000
low_veg
0.1791
0.0967
0.5067
0.2109
0.0067
0.0000
tree
0.0603
0.0167
0.0708
0.8493
0.0029
0.0000
car
0.2399
0.0387
0.0103
0.0028
0.7084
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.862
0.938
0.661
0.768
0.510
--
Recall/Completeness
0.853
0.943
0.507
0.849
0.708
--
F1
0.857
0.941
0.574
0.807
0.593
--

Overall accuracy 0.887

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
0.8903
0.0739
0.0107
0.0030
0.0221
0.0000
building
0.0336
0.9577
0.0060
0.0011
0.0016
0.0000
low_veg
0.1487
0.0588
0.5607
0.2288
0.0030
0.0000
tree
0.0358
0.0066
0.0637
0.8929
0.0010
0.0000
car
0.2103
0.0307
0.0077
0.0014
0.7499
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.895
0.959
0.699
0.797
0.527
--
Recall/Completeness
0.890
0.958
0.561
0.893
0.750
--
F1
0.893
0.958
0.622
0.842
0.619
--

Overall accuracy 0.916

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
0.8388
0.0751
0.0428
0.0103
0.0330
0.0000
building
0.0350
0.9386
0.0175
0.0071
0.0019
0.0000
low_veg
0.0578
0.0446
0.7312
0.1630
0.0033
0.0000
tree
0.0119
0.0048
0.0655
0.9177
0.0002
0.0000
car
0.2203
0.0362
0.0096
0.0032
0.7306
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.896
0.923
0.686
0.922
0.558
--
Recall/Completeness
0.839
0.939
0.731
0.918
0.731
--
F1
0.867
0.931
0.708
0.920
0.633
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8864
0.0557
0.0314
0.0035
0.0231
0.0000
building
0.0274
0.9554
0.0119
0.0043
0.0010
0.0000
low_veg
0.0458
0.0294
0.7801
0.1421
0.0025
0.0000
tree
0.0046
0.0024
0.0510
0.9420
0.0000
0.0000
car
0.1845
0.0293
0.0072
0.0008
0.7782
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.927
0.948
0.736
0.945
0.601
--
Recall/Completeness
0.886
0.955
0.780
0.942
0.778
--
F1
0.906
0.952
0.758
0.944
0.678
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8531
0.0363
0.0712
0.0144
0.0251
0.0000
building
0.0309
0.9471
0.0152
0.0049
0.0019
0.0000
low_veg
0.0840
0.0183
0.8471
0.0486
0.0021
0.0000
tree
0.0257
0.0083
0.1046
0.8604
0.0010
0.0000
car
0.1919
0.0078
0.0295
0.0059
0.7649
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.890
0.928
0.764
0.931
0.592
--
Recall/Completeness
0.853
0.947
0.847
0.860
0.765
--
F1
0.871
0.937
0.804
0.894
0.667
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8874
0.0294
0.0600
0.0061
0.0170
0.0000
building
0.0203
0.9665
0.0094
0.0024
0.0014
0.0000
low_veg
0.0687
0.0098
0.8824
0.0382
0.0009
0.0000
tree
0.0136
0.0042
0.0864
0.8953
0.0005
0.0000
car
0.1528
0.0051
0.0248
0.0043
0.8129
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.922
0.950
0.803
0.956
0.636
--
Recall/Completeness
0.887
0.966
0.882
0.895
0.813
--
F1
0.904
0.958
0.841
0.925
0.714
--

Overall accuracy 0.906

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8418
0.0707
0.0525
0.0215
0.0134
0.0000
building
0.0146
0.9496
0.0195
0.0147
0.0016
0.0000
low_veg
0.0680
0.0329
0.7387
0.1578
0.0026
0.0000
tree
0.0174
0.0076
0.0938
0.8810
0.0002
0.0000
car
0.1505
0.0398
0.0191
0.0083
0.7823
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.940
0.868
0.748
0.822
0.483
--
Recall/Completeness
0.842
0.950
0.739
0.881
0.782
--
F1
0.888
0.907
0.743
0.850
0.597
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8801
0.0567
0.0427
0.0106
0.0100
0.0000
building
0.0096
0.9648
0.0130
0.0115
0.0011
0.0000
low_veg
0.0636
0.0202
0.7843
0.1294
0.0025
0.0000
tree
0.0094
0.0039
0.0741
0.9126
0.0000
0.0000
car
0.1197
0.0419
0.0192
0.0041
0.8152
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.956
0.899
0.794
0.864
0.486
--
Recall/Completeness
0.880
0.965
0.784
0.913
0.815
--
F1
0.916
0.931
0.789
0.888
0.609
--

Overall accuracy 0.890

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8703
0.0331
0.0401
0.0436
0.0128
0.0000
building
0.1813
0.8026
0.0091
0.0069
0.0000
0.0000
low_veg
0.0576
0.0332
0.6815
0.2262
0.0014
0.0000
tree
0.0117
0.0091
0.1014
0.8776
0.0001
0.0000
car
0.1006
0.0005
0.0003
0.0010
0.8976
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.688
0.896
0.815
0.816
0.662
--
Recall/Completeness
0.870
0.803
0.682
0.878
0.898
--
F1
0.768
0.847
0.742
0.846
0.762
--

Overall accuracy 0.805

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
0.9021
0.0299
0.0321
0.0282
0.0077
0.0000
building
0.1879
0.8065
0.0028
0.0028
0.0000
0.0000
low_veg
0.0543
0.0225
0.7158
0.2061
0.0013
0.0000
tree
0.0066
0.0048
0.0861
0.9025
0.0000
0.0000
car
0.0292
0.0000
0.0000
0.0001
0.9707
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.703
0.927
0.845
0.842
0.706
--
Recall/Completeness
0.902
0.807
0.716
0.902
0.971
--
F1
0.790
0.862
0.775
0.871
0.818
--

Overall accuracy 0.831

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
0.8545
0.0374
0.0682
0.0312
0.0087
0.0000
building
0.0228
0.9549
0.0129
0.0081
0.0013
0.0000
low_veg
0.0832
0.0584
0.7140
0.1427
0.0016
0.0000
tree
0.0317
0.0143
0.1156
0.8378
0.0005
0.0000
car
0.0930
0.0011
0.0008
0.0072
0.8979
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.845
0.856
0.762
0.862
0.460
--
Recall/Completeness
0.854
0.955
0.714
0.838
0.898
--
F1
0.850
0.903
0.737
0.850
0.609
--

Overall accuracy 0.831

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
0.8789
0.0319
0.0615
0.0213
0.0064
0.0000
building
0.0153
0.9735
0.0061
0.0041
0.0010
0.0000
low_veg
0.0768
0.0431
0.7490
0.1297
0.0014
0.0000
tree
0.0240
0.0083
0.1035
0.8639
0.0004
0.0000
car
0.0500
0.0008
0.0001
0.0031
0.9460
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.872
0.890
0.787
0.885
0.465
--
Recall/Completeness
0.879
0.973
0.749
0.864
0.946
--
F1
0.876
0.930
0.768
0.874
0.623
--

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8137
0.0259
0.0807
0.0688
0.0109
0.0000
building
0.0346
0.9325
0.0219
0.0108
0.0002
0.0000
low_veg
0.0321
0.0202
0.7487
0.1984
0.0006
0.0000
tree
0.0224
0.0086
0.0916
0.8772
0.0002
0.0000
car
0.0957
0.0008
0.0050
0.0032
0.8954
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.839
0.927
0.816
0.818
0.502
--
Recall/Completeness
0.814
0.933
0.749
0.877
0.895
--
F1
0.826
0.930
0.781
0.847
0.644
--

Overall accuracy 0.839

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
0.8501
0.0204
0.0741
0.0477
0.0077
0.0000
building
0.0290
0.9515
0.0145
0.0049
0.0001
0.0000
low_veg
0.0276
0.0122
0.7757
0.1839
0.0006
0.0000
tree
0.0156
0.0044
0.0808
0.8991
0.0001
0.0000
car
0.0444
0.0000
0.0000
0.0000
0.9556
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.870
0.953
0.841
0.841
0.521
--
Recall/Completeness
0.850
0.951
0.776
0.899
0.956
--
F1
0.860
0.952
0.807
0.869
0.674
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8209
0.0352
0.0900
0.0399
0.0140
0.0000
building
0.0133
0.9419
0.0319
0.0126
0.0003
0.0000
low_veg
0.0432
0.0231
0.7859
0.1461
0.0017
0.0000
tree
0.0092
0.0018
0.0716
0.9173
0.0001
0.0000
car
0.1645
0.0050
0.0103
0.0109
0.8093
0.0000
clutter
0.0194
0.8053
0.1725
0.0028
0.0000
0.0000
Precision/Correctness
0.914
0.819
0.726
0.907
0.509
-nan
Recall/Completeness
0.821
0.942
0.786
0.917
0.809
0.000
F1
0.865
0.876
0.755
0.912
0.625
-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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8553
0.0268
0.0829
0.0241
0.0109
0.0000
building
0.0107
0.9556
0.0259
0.0077
0.0001
0.0000
low_veg
0.0371
0.0125
0.8306
0.1183
0.0015
0.0000
tree
0.0047
0.0005
0.0564
0.9384
0.0000
0.0000
car
0.1166
0.0011
0.0027
0.0068
0.8728
0.0000
clutter
0.0145
0.8265
0.1576
0.0014
0.0000
0.0000
Precision/Correctness
0.938
0.836
0.758
0.933
0.508
-nan
Recall/Completeness
0.855
0.956
0.831
0.938
0.873
0.000
F1
0.895
0.892
0.793
0.936
0.642
-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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8681
0.0272
0.0501
0.0361
0.0185
0.0000
building
0.0341
0.9430
0.0146
0.0080
0.0003
0.0000
low_veg
0.0928
0.0406
0.7053
0.1608
0.0004
0.0000
tree
0.0233
0.0056
0.0781
0.8922
0.0008
0.0000
car
0.2853
0.0066
0.0331
0.0062
0.6688
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.889
0.900
0.774
0.850
0.539
--
Recall/Completeness
0.868
0.943
0.705
0.892
0.669
--
F1
0.878
0.921
0.738
0.871
0.597
--

Overall accuracy 0.853

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
0.9031
0.0206
0.0410
0.0210
0.0143
0.0000
building
0.0250
0.9630
0.0077
0.0042
0.0001
0.0000
low_veg
0.0897
0.0262
0.7474
0.1367
0.0001
0.0000
tree
0.0127
0.0025
0.0609
0.9236
0.0003
0.0000
car
0.2455
0.0096
0.0250
0.0018
0.7180
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.914
0.933
0.818
0.886
0.550
--
Recall/Completeness
0.903
0.963
0.747
0.924
0.718
--
F1
0.908
0.948
0.781
0.904
0.623
--

Overall accuracy 0.888

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7903
0.0481
0.0982
0.0459
0.0175
0.0000
building
0.0648
0.8924
0.0269
0.0146
0.0013
0.0000
low_veg
0.0275
0.0211
0.8460
0.1048
0.0005
0.0000
tree
0.0162
0.0112
0.1364
0.8359
0.0002
0.0000
car
0.1598
0.0000
0.0138
0.0023
0.8241
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.803
0.901
0.866
0.802
0.504
--
Recall/Completeness
0.790
0.892
0.846
0.836
0.824
--
F1
0.797
0.896
0.856
0.818
0.626
--

Overall accuracy 0.844

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
0.8272
0.0391
0.0876
0.0334
0.0127
0.0000
building
0.0602
0.9061
0.0213
0.0111
0.0014
0.0000
low_veg
0.0243
0.0161
0.8685
0.0906
0.0005
0.0000
tree
0.0102
0.0070
0.1219
0.8607
0.0001
0.0000
car
0.1199
0.0000
0.0073
0.0001
0.8727
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.828
0.925
0.885
0.828
0.504
--
Recall/Completeness
0.827
0.906
0.869
0.861
0.873
--
F1
0.828
0.915
0.877
0.844
0.639
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8871
0.0316
0.0246
0.0078
0.0489
0.0000
building
0.0556
0.9217
0.0105
0.0099
0.0023
0.0000
low_veg
0.1079
0.0325
0.7063
0.1494
0.0039
0.0000
tree
0.0147
0.0086
0.0665
0.9095
0.0006
0.0000
car
0.1884
0.0069
0.0095
0.0062
0.7889
0.0000
clutter
0.6476
0.0811
0.0909
0.0087
0.1717
0.0000
Precision/Correctness
0.851
0.948
0.881
0.720
0.592
-nan
Recall/Completeness
0.887
0.922
0.706
0.910
0.789
0.000
F1
0.868
0.935
0.784
0.804
0.677
-nan

Overall accuracy 0.861

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
0.9278
0.0210
0.0168
0.0038
0.0306
0.0000
building
0.0434
0.9414
0.0058
0.0079
0.0016
0.0000
low_veg
0.0929
0.0176
0.7505
0.1362
0.0028
0.0000
tree
0.0072
0.0042
0.0497
0.9388
0.0001
0.0000
car
0.1569
0.0040
0.0073
0.0052
0.8267
0.0000
clutter
0.6546
0.0766
0.0700
0.0044
0.1943
0.0000
Precision/Correctness
0.883
0.970
0.918
0.754
0.644
-nan
Recall/Completeness
0.928
0.941
0.751
0.939
0.827
0.000
F1
0.905
0.955
0.826
0.836
0.724
-nan

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9156
0.0222
0.0235
0.0146
0.0242
0.0000
building
0.0681
0.9086
0.0076
0.0083
0.0073
0.0000
low_veg
0.0571
0.0363
0.6162
0.2854
0.0050
0.0000
tree
0.0110
0.0034
0.0267
0.9588
0.0001
0.0000
car
0.1287
0.0067
0.0008
0.0114
0.8524
0.0000
clutter
0.4041
0.2618
0.0371
0.0081
0.2889
0.0000
Precision/Correctness
0.903
0.915
0.850
0.673
0.346
-nan
Recall/Completeness
0.916
0.909
0.616
0.959
0.852
0.000
F1
0.909
0.912
0.715
0.791
0.493
-nan

Overall accuracy 0.846

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
0.9426
0.0144
0.0171
0.0078
0.0181
0.0000
building
0.0657
0.9183
0.0039
0.0049
0.0073
0.0000
low_veg
0.0384
0.0198
0.6653
0.2731
0.0034
0.0000
tree
0.0060
0.0013
0.0201
0.9726
0.0000
0.0000
car
0.0916
0.0032
0.0001
0.0086
0.8965
0.0000
clutter
0.3848
0.2657
0.0319
0.0021
0.3155
0.0000
Precision/Correctness
0.919
0.941
0.888
0.710
0.327
-nan
Recall/Completeness
0.943
0.918
0.665
0.973
0.897
0.000
F1
0.931
0.929
0.761
0.821
0.479
-nan

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9234
0.0128
0.0102
0.0074
0.0461
0.0000
building
0.2366
0.7419
0.0084
0.0022
0.0110
0.0000
low_veg
0.0512
0.0051
0.9138
0.0254
0.0044
0.0000
tree
0.0256
0.0056
0.0892
0.8733
0.0063
0.0000
car
0.0723
0.0826
0.0065
0.0047
0.8340
0.0000
clutter
0.4533
0.3630
0.0195
0.0161
0.1481
0.0000
Precision/Correctness
0.793
0.905
0.887
0.925
0.523
-nan
Recall/Completeness
0.923
0.742
0.914
0.873
0.834
0.000
F1
0.853
0.815
0.900
0.899
0.643
-nan

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9494
0.0090
0.0078
0.0028
0.0309
0.0000
building
0.2326
0.7535
0.0042
0.0004
0.0093
0.0000
low_veg
0.0305
0.0028
0.9476
0.0171
0.0020
0.0000
tree
0.0087
0.0019
0.0739
0.9127
0.0028
0.0000
car
0.0393
0.0962
0.0026
0.0014
0.8606
0.0000
clutter
0.4136
0.4151
0.0133
0.0081
0.1499
0.0000
Precision/Correctness
0.810
0.918
0.917
0.961
0.554
-nan
Recall/Completeness
0.949
0.754
0.948
0.913
0.861
0.000
F1
0.874
0.828
0.932
0.936
0.674
-nan

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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9548
0.0121
0.0097
0.0032
0.0203
0.0000
building
0.2037
0.7166
0.0235
0.0508
0.0054
0.0000
low_veg
0.0810
0.0161
0.7766
0.1239
0.0024
0.0000
tree
0.0401
0.0123
0.1814
0.7661
0.0001
0.0000
car
0.2164
0.0099
0.0185
0.0019
0.7533
0.0000
clutter
0.4668
0.1051
0.0050
0.0004
0.4226
0.0000
Precision/Correctness
0.780
0.936
0.799
0.696
0.229
-nan
Recall/Completeness
0.955
0.717
0.777
0.766
0.753
0.000
F1
0.859
0.812
0.788
0.729
0.351
-nan

Overall accuracy 0.784

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
0.9720
0.0082
0.0051
0.0011
0.0136
0.0000
building
0.2018
0.7223
0.0194
0.0514
0.0050
0.0000
low_veg
0.0512
0.0113
0.8045
0.1318
0.0012
0.0000
tree
0.0173
0.0068
0.1683
0.8076
0.0000
0.0000
car
0.1759
0.0079
0.0172
0.0008
0.7981
0.0000
clutter
0.4367
0.1036
0.0027
0.0000
0.4570
0.0000
Precision/Correctness
0.799
0.953
0.828
0.698
0.210
-nan
Recall/Completeness
0.972
0.722
0.805
0.808
0.798
0.000
F1
0.877
0.822
0.816
0.749
0.333
-nan

Overall accuracy 0.804

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
0.8353
0.0394
0.0875
0.0240
0.0137
0.0000
building
0.0379
0.9269
0.0252
0.0058
0.0043
0.0000
low_veg
0.0506
0.0156
0.7296
0.2033
0.0009
0.0000
tree
0.0266
0.0057
0.1328
0.8345
0.0004
0.0000
car
0.2110
0.0214
0.0067
0.0145
0.7465
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.873
0.942
0.752
0.771
0.535
--
Recall/Completeness
0.835
0.927
0.730
0.835
0.746
--
F1
0.854
0.934
0.740
0.801
0.623
--

Overall accuracy 0.833

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
0.8878
0.0343
0.0590
0.0092
0.0098
0.0000
building
0.0251
0.9568
0.0128
0.0018
0.0035
0.0000
low_veg
0.0397
0.0081
0.7577
0.1939
0.0007
0.0000
tree
0.0112
0.0018
0.1047
0.8822
0.0001
0.0000
car
0.1546
0.0281
0.0051
0.0071
0.8051
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.915
0.961
0.815
0.800
0.569
--
Recall/Completeness
0.888
0.957
0.758
0.882
0.805
--
F1
0.901
0.959
0.785
0.839
0.667
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8748
0.0335
0.0500
0.0214
0.0202
0.0000
building
0.0213
0.9269
0.0405
0.0075
0.0037
0.0000
low_veg
0.0403
0.0199
0.8510
0.0869
0.0019
0.0000
tree
0.0238
0.0056
0.1275
0.8423
0.0008
0.0000
car
0.1370
0.0168
0.0702
0.0113
0.7647
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.912
0.937
0.829
0.837
0.511
--
Recall/Completeness
0.875
0.927
0.851
0.842
0.765
--
F1
0.893
0.932
0.840
0.839
0.613
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9089
0.0294
0.0362
0.0107
0.0148
0.0000
building
0.0134
0.9503
0.0293
0.0039
0.0031
0.0000
low_veg
0.0259
0.0099
0.8920
0.0709
0.0013
0.0000
tree
0.0143
0.0021
0.1033
0.8799
0.0004
0.0000
car
0.0926
0.0161
0.0700
0.0053
0.8160
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.945
0.957
0.868
0.876
0.531
--
Recall/Completeness
0.909
0.950
0.892
0.880
0.816
--
F1
0.927
0.954
0.880
0.878
0.644
--

Overall accuracy 0.909

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image