Detailled semantic labeling (2D) result


Name D. Tran and T. Bui
Affiliation University​ ​of​ ​Engineering​ ​and​ ​Technology, Vietnam
Abbreviation VNU4
Strategy (u)nsupervised, (s)upervised, (h)ybrid s


Overall statistics, reference set: full_reference


Number of non processed tiles by this participant: 2
↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.909
0.033
0.039
0.015
0.004
0.000
building
0.062
0.908
0.024
0.005
0.001
0.000
low_veg
0.061
0.025
0.761
0.153
0.000
0.000
tree
0.024
0.005
0.099
0.872
0.000
0.000
car
0.301
0.048
0.010
0.006
0.632
0.003
clutter
0.348
0.299
0.017
0.009
0.060
0.268
Precision/Correctness
0.860
0.927
0.799
0.839
0.797
0.949
Recall/Completeness
0.909
0.908
0.761
0.872
0.632
0.268
F1
0.884
0.918
0.779
0.855
0.705
0.418

Overall accuracy 0.860

Overall statistics, reference set: no_boundary


Number of non processed tiles by this participant: 2
↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.934
0.026
0.031
0.007
0.003
0.000
building
0.054
0.926
0.017
0.002
0.001
0.000
low_veg
0.048
0.016
0.797
0.139
0.000
0.000
tree
0.012
0.002
0.083
0.903
0.000
0.000
car
0.201
0.053
0.006
0.003
0.734
0.003
clutter
0.307
0.319
0.015
0.003
0.062
0.294
Precision/Correctness
0.891
0.947
0.834
0.868
0.825
0.965
Recall/Completeness
0.934
0.926
0.797
0.903
0.734
0.294
F1
0.912
0.936
0.815
0.885
0.777
0.451

Overall accuracy 0.890
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9419
0.0144
0.0256
0.0132
0.0049
0.0000
building
0.0461
0.9269
0.0229
0.0037
0.0004
0.0000
low_veg
0.0766
0.0213
0.7231
0.1788
0.0002
0.0000
tree
0.0309
0.0042
0.1114
0.8521
0.0001
0.0013
car
0.3116
0.0112
0.0006
0.0019
0.6746
0.0000
clutter
0.0076
0.0030
0.0023
0.0188
0.0014
0.9669
Precision/Correctness
0.849
0.955
0.819
0.813
0.881
0.982
Recall/Completeness
0.942
0.927
0.723
0.852
0.675
0.967
F1
0.893
0.941
0.768
0.832
0.764
0.974

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9657
0.0108
0.0163
0.0039
0.0033
0.0000
building
0.0378
0.9472
0.0132
0.0014
0.0004
0.0000
low_veg
0.0591
0.0122
0.7555
0.1730
0.0001
0.0000
tree
0.0152
0.0015
0.0992
0.8835
0.0000
0.0006
car
0.1913
0.0105
0.0000
0.0008
0.7974
0.0000
clutter
0.0078
0.0025
0.0024
0.0049
0.0012
0.9812
Precision/Correctness
0.890
0.973
0.850
0.837
0.909
0.992
Recall/Completeness
0.966
0.947
0.756
0.884
0.797
0.981
F1
0.926
0.960
0.800
0.860
0.850
0.986

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9389
0.0428
0.0100
0.0037
0.0047
0.0000
building
0.0903
0.8979
0.0088
0.0020
0.0010
0.0000
low_veg
0.2052
0.0430
0.4752
0.2763
0.0003
0.0000
tree
0.0593
0.0149
0.0982
0.8275
0.0001
0.0000
car
0.4375
0.0137
0.0014
0.0025
0.5449
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.799
0.971
0.627
0.732
0.801
--
Recall/Completeness
0.939
0.898
0.475
0.827
0.545
--
F1
0.863
0.933
0.541
0.777
0.649
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9647
0.0239
0.0069
0.0013
0.0032
0.0000
building
0.0772
0.9130
0.0076
0.0011
0.0010
0.0000
low_veg
0.1688
0.0148
0.5100
0.3064
0.0001
0.0000
tree
0.0277
0.0066
0.0965
0.8691
0.0000
0.0000
car
0.3499
0.0072
0.0004
0.0016
0.6410
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.827
0.986
0.650
0.754
0.823
--
Recall/Completeness
0.965
0.913
0.510
0.869
0.641
--
F1
0.891
0.948
0.572
0.808
0.721
--

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_area6_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9079
0.0463
0.0320
0.0099
0.0039
0.0000
building
0.0710
0.9108
0.0142
0.0035
0.0005
0.0000
low_veg
0.1034
0.0318
0.6951
0.1692
0.0005
0.0000
tree
0.0177
0.0029
0.0559
0.9232
0.0002
0.0000
car
0.4001
0.0140
0.0023
0.0018
0.5818
0.0001
clutter
--
--
--
--
--
--
Precision/Correctness
0.834
0.949
0.719
0.924
0.883
--
Recall/Completeness
0.908
0.911
0.695
0.923
0.582
--
F1
0.869
0.929
0.707
0.924
0.701
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9408
0.0310
0.0236
0.0027
0.0019
0.0000
building
0.0620
0.9263
0.0099
0.0014
0.0004
0.0000
low_veg
0.0855
0.0177
0.7460
0.1507
0.0002
0.0000
tree
0.0069
0.0006
0.0436
0.9488
0.0001
0.0000
car
0.2879
0.0083
0.0012
0.0001
0.7025
0.0001
clutter
--
--
--
--
--
--
Precision/Correctness
0.870
0.970
0.765
0.947
0.928
--
Recall/Completeness
0.941
0.926
0.746
0.949
0.702
--
F1
0.904
0.948
0.756
0.948
0.800
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8921
0.0249
0.0699
0.0095
0.0035
0.0000
building
0.0377
0.9378
0.0182
0.0056
0.0007
0.0000
low_veg
0.0558
0.0137
0.8640
0.0661
0.0004
0.0000
tree
0.0262
0.0058
0.0884
0.8796
0.0001
0.0000
car
0.3124
0.0070
0.0251
0.0034
0.6504
0.0018
clutter
--
--
--
--
--
--
Precision/Correctness
0.900
0.947
0.781
0.924
0.890
--
Recall/Completeness
0.892
0.938
0.864
0.880
0.650
--
F1
0.896
0.943
0.820
0.901
0.752
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9188
0.0180
0.0593
0.0021
0.0017
0.0000
building
0.0283
0.9578
0.0104
0.0029
0.0006
0.0000
low_veg
0.0398
0.0059
0.8995
0.0547
0.0001
0.0000
tree
0.0112
0.0024
0.0683
0.9181
0.0000
0.0000
car
0.2022
0.0044
0.0160
0.0020
0.7729
0.0025
clutter
--
--
--
--
--
--
Precision/Correctness
0.936
0.969
0.823
0.948
0.933
--
Recall/Completeness
0.919
0.958
0.899
0.918
0.773
--
F1
0.927
0.963
0.859
0.933
0.846
--

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_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8325
0.0820
0.0634
0.0184
0.0037
0.0000
building
0.0380
0.9307
0.0204
0.0102
0.0006
0.0001
low_veg
0.0800
0.0346
0.7264
0.1576
0.0012
0.0001
tree
0.0170
0.0069
0.0789
0.8972
0.0000
0.0000
car
0.2569
0.0376
0.0122
0.0067
0.6866
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.917
0.851
0.740
0.833
0.734
--
Recall/Completeness
0.833
0.931
0.726
0.897
0.687
--
F1
0.873
0.889
0.733
0.864
0.709
--

Overall accuracy 0.849

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
0.8640
0.0702
0.0558
0.0073
0.0027
0.0000
building
0.0321
0.9453
0.0151
0.0069
0.0004
0.0001
low_veg
0.0769
0.0214
0.7670
0.1332
0.0013
0.0001
tree
0.0077
0.0030
0.0586
0.9307
0.0000
0.0000
car
0.1670
0.0431
0.0082
0.0022
0.7796
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.935
0.879
0.779
0.873
0.752
--
Recall/Completeness
0.864
0.945
0.767
0.931
0.780
--
F1
0.898
0.911
0.773
0.901
0.765
--

Overall accuracy 0.880

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8930
0.0288
0.0419
0.0337
0.0026
0.0000
building
0.1280
0.8520
0.0142
0.0057
0.0000
0.0000
low_veg
0.0495
0.0297
0.7419
0.1784
0.0005
0.0000
tree
0.0114
0.0077
0.1175
0.8630
0.0001
0.0002
car
0.2186
0.0034
0.0067
0.0030
0.7684
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.742
0.913
0.805
0.847
0.874
--
Recall/Completeness
0.893
0.852
0.742
0.863
0.768
--
F1
0.810
0.881
0.772
0.855
0.818
--

Overall accuracy 0.830

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
0.9208
0.0258
0.0346
0.0182
0.0006
0.0000
building
0.1329
0.8586
0.0066
0.0019
0.0000
0.0000
low_veg
0.0466
0.0203
0.7753
0.1574
0.0004
0.0000
tree
0.0058
0.0029
0.1018
0.8892
0.0001
0.0002
car
0.1046
0.0017
0.0037
0.0000
0.8899
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.760
0.941
0.833
0.874
0.932
--
Recall/Completeness
0.921
0.859
0.775
0.889
0.890
--
F1
0.833
0.898
0.803
0.882
0.911
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8356
0.0321
0.0944
0.0338
0.0041
0.0000
building
0.0131
0.9612
0.0194
0.0062
0.0001
0.0000
low_veg
0.0811
0.0501
0.7082
0.1603
0.0002
0.0000
tree
0.0316
0.0118
0.1252
0.8311
0.0001
0.0001
car
0.1651
0.0000
0.0003
0.0035
0.8311
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.850
0.875
0.729
0.849
0.681
--
Recall/Completeness
0.836
0.961
0.708
0.831
0.831
--
F1
0.843
0.916
0.718
0.840
0.749
--

Overall accuracy 0.824

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
0.8554
0.0267
0.0914
0.0232
0.0033
0.0000
building
0.0080
0.9792
0.0102
0.0025
0.0001
0.0000
low_veg
0.0767
0.0346
0.7378
0.1508
0.0002
0.0000
tree
0.0232
0.0054
0.1132
0.8581
0.0000
0.0001
car
0.0497
0.0000
0.0000
0.0002
0.9500
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.875
0.911
0.749
0.870
0.693
--
Recall/Completeness
0.855
0.979
0.738
0.858
0.950
--
F1
0.865
0.944
0.744
0.864
0.801
--

Overall accuracy 0.849

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
0.8367
0.0305
0.0806
0.0497
0.0025
0.0000
building
0.0210
0.9485
0.0220
0.0082
0.0003
0.0000
low_veg
0.0224
0.0203
0.8409
0.1153
0.0004
0.0007
tree
0.0265
0.0038
0.1359
0.8337
0.0000
0.0000
car
0.2187
0.0004
0.0074
0.0054
0.7681
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.863
0.933
0.790
0.876
0.749
--
Recall/Completeness
0.837
0.949
0.841
0.834
0.768
--
F1
0.849
0.941
0.815
0.854
0.758
--

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_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8698
0.0263
0.0726
0.0304
0.0009
0.0000
building
0.0164
0.9699
0.0111
0.0024
0.0002
0.0000
low_veg
0.0184
0.0142
0.8651
0.1012
0.0004
0.0007
tree
0.0184
0.0010
0.1210
0.8596
0.0000
0.0000
car
0.1407
0.0000
0.0032
0.0010
0.8552
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.895
0.953
0.817
0.900
0.820
--
Recall/Completeness
0.870
0.970
0.865
0.860
0.855
--
F1
0.882
0.961
0.841
0.879
0.837
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8530
0.0557
0.0399
0.0462
0.0052
0.0001
building
0.0243
0.9437
0.0261
0.0059
0.0000
0.0000
low_veg
0.0521
0.0348
0.7627
0.1504
0.0000
0.0000
tree
0.0094
0.0020
0.0545
0.9340
0.0000
0.0000
car
0.2415
0.0052
0.0400
0.0133
0.6999
0.0000
clutter
0.0298
0.9695
0.0000
0.0007
0.0000
0.0000
Precision/Correctness
0.897
0.774
0.809
0.907
0.734
0.000
Recall/Completeness
0.853
0.944
0.763
0.934
0.700
0.000
F1
0.874
0.851
0.785
0.920
0.716
-nan

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8816
0.0483
0.0378
0.0285
0.0037
0.0001
building
0.0214
0.9549
0.0218
0.0019
0.0000
0.0000
low_veg
0.0459
0.0217
0.8087
0.1236
0.0000
0.0000
tree
0.0044
0.0005
0.0417
0.9534
0.0000
0.0000
car
0.1796
0.0031
0.0358
0.0051
0.7764
0.0000
clutter
0.0224
0.9776
0.0000
0.0000
0.0000
0.0000
Precision/Correctness
0.922
0.793
0.839
0.932
0.752
0.000
Recall/Completeness
0.882
0.955
0.809
0.953
0.776
0.000
F1
0.901
0.866
0.823
0.943
0.764
-nan

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9361
0.0277
0.0215
0.0075
0.0072
0.0000
building
0.0898
0.8804
0.0239
0.0052
0.0005
0.0001
low_veg
0.0937
0.0232
0.6710
0.2117
0.0004
0.0000
tree
0.0142
0.0027
0.0593
0.9236
0.0001
0.0000
car
0.3290
0.0097
0.0055
0.0062
0.6494
0.0001
clutter
0.7133
0.1290
0.0884
0.0031
0.0663
0.0000
Precision/Correctness
0.825
0.956
0.859
0.669
0.886
0.000
Recall/Completeness
0.936
0.880
0.671
0.924
0.649
0.000
F1
0.877
0.916
0.753
0.776
0.750
-nan

Overall accuracy 0.852

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
0.9618
0.0189
0.0142
0.0028
0.0023
0.0000
building
0.0779
0.9009
0.0179
0.0029
0.0004
0.0001
low_veg
0.0770
0.0116
0.7096
0.2016
0.0002
0.0000
tree
0.0059
0.0004
0.0412
0.9525
0.0000
0.0000
car
0.2260
0.0077
0.0024
0.0047
0.7590
0.0002
clutter
0.7386
0.1248
0.0620
0.0011
0.0735
0.0000
Precision/Correctness
0.858
0.974
0.898
0.698
0.946
0.000
Recall/Completeness
0.962
0.901
0.710
0.952
0.759
0.000
F1
0.907
0.936
0.793
0.806
0.842
-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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9312
0.0261
0.0245
0.0128
0.0051
0.0002
building
0.0530
0.9262
0.0143
0.0062
0.0003
0.0000
low_veg
0.0798
0.0204
0.5894
0.3096
0.0008
0.0000
tree
0.0064
0.0015
0.0295
0.9625
0.0000
0.0000
car
0.2422
0.0116
0.0010
0.0167
0.6991
0.0293
clutter
0.5089
0.3710
0.0150
0.0157
0.0702
0.0192
Precision/Correctness
0.897
0.908
0.829
0.663
0.679
0.539
Recall/Completeness
0.931
0.926
0.589
0.963
0.699
0.019
F1
0.914
0.917
0.689
0.785
0.689
0.037

Overall accuracy 0.852

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
0.9520
0.0205
0.0182
0.0055
0.0036
0.0001
building
0.0472
0.9406
0.0090
0.0030
0.0002
0.0000
low_veg
0.0568
0.0086
0.6349
0.2991
0.0006
0.0000
tree
0.0021
0.0002
0.0222
0.9755
0.0000
0.0000
car
0.1464
0.0094
0.0002
0.0146
0.7961
0.0333
clutter
0.4870
0.3985
0.0110
0.0063
0.0756
0.0216
Precision/Correctness
0.920
0.927
0.869
0.700
0.685
0.601
Recall/Completeness
0.952
0.941
0.635
0.976
0.796
0.022
F1
0.936
0.934
0.734
0.815
0.737
0.042

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9648
0.0143
0.0115
0.0030
0.0064
0.0001
building
0.0976
0.8865
0.0109
0.0021
0.0029
0.0001
low_veg
0.0413
0.0055
0.9192
0.0333
0.0006
0.0000
tree
0.0393
0.0028
0.1009
0.8565
0.0005
0.0000
car
0.1990
0.2064
0.0089
0.0052
0.5769
0.0036
clutter
0.5120
0.2471
0.0509
0.0120
0.1202
0.0578
Precision/Correctness
0.858
0.916
0.867
0.930
0.735
0.920
Recall/Completeness
0.965
0.886
0.919
0.857
0.577
0.058
F1
0.908
0.901
0.892
0.892
0.646
0.109

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9767
0.0114
0.0082
0.0004
0.0034
0.0000
building
0.0875
0.9056
0.0042
0.0004
0.0023
0.0000
low_veg
0.0227
0.0033
0.9455
0.0282
0.0003
0.0000
tree
0.0127
0.0009
0.0810
0.9054
0.0001
0.0000
car
0.1208
0.2318
0.0024
0.0014
0.6396
0.0041
clutter
0.4626
0.2827
0.0536
0.0039
0.1299
0.0673
Precision/Correctness
0.889
0.929
0.904
0.955
0.756
0.938
Recall/Completeness
0.977
0.906
0.946
0.905
0.640
0.067
F1
0.931
0.917
0.924
0.930
0.693
0.126

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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9693
0.0173
0.0070
0.0016
0.0029
0.0019
building
0.0626
0.8475
0.0845
0.0032
0.0022
0.0000
low_veg
0.0679
0.0188
0.8294
0.0834
0.0006
0.0000
tree
0.0527
0.0040
0.1381
0.8050
0.0002
0.0000
car
0.3629
0.0556
0.0079
0.0048
0.5688
0.0000
clutter
0.4492
0.1136
0.0024
0.0009
0.0710
0.3629
Precision/Correctness
0.855
0.936
0.753
0.853
0.561
0.964
Recall/Completeness
0.969
0.848
0.829
0.805
0.569
0.363
F1
0.909
0.890
0.789
0.828
0.565
0.527

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9805
0.0133
0.0029
0.0003
0.0023
0.0008
building
0.0547
0.8585
0.0833
0.0014
0.0021
0.0000
low_veg
0.0383
0.0101
0.8666
0.0845
0.0006
0.0000
tree
0.0250
0.0011
0.1244
0.8493
0.0002
0.0000
car
0.2500
0.0632
0.0060
0.0018
0.6790
0.0000
clutter
0.4116
0.1126
0.0014
0.0005
0.0771
0.3967
Precision/Correctness
0.887
0.952
0.771
0.867
0.551
0.985
Recall/Completeness
0.980
0.859
0.867
0.849
0.679
0.397
F1
0.932
0.903
0.816
0.858
0.608
0.566

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8622
0.0465
0.0712
0.0187
0.0014
0.0000
building
0.0580
0.9099
0.0259
0.0054
0.0008
0.0000
low_veg
0.0429
0.0141
0.7096
0.2334
0.0000
0.0000
tree
0.0334
0.0050
0.1153
0.8460
0.0003
0.0000
car
0.3493
0.0191
0.0069
0.0088
0.6157
0.0002
clutter
--
--
--
--
--
--
Precision/Correctness
0.855
0.936
0.773
0.756
0.880
--
Recall/Completeness
0.862
0.910
0.710
0.846
0.616
--
F1
0.859
0.923
0.740
0.798
0.724
--

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_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9059
0.0399
0.0492
0.0045
0.0005
0.0000
building
0.0476
0.9353
0.0152
0.0011
0.0008
0.0000
low_veg
0.0301
0.0063
0.7362
0.2275
0.0000
0.0000
tree
0.0136
0.0015
0.0893
0.8955
0.0001
0.0000
car
0.2302
0.0224
0.0031
0.0026
0.7413
0.0003
clutter
--
--
--
--
--
--
Precision/Correctness
0.898
0.957
0.830
0.782
0.921
--
Recall/Completeness
0.906
0.935
0.736
0.896
0.741
--
F1
0.902
0.946
0.780
0.835
0.822
--

Overall accuracy 0.870

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
0.9199
0.0246
0.0399
0.0120
0.0036
0.0000
building
0.0364
0.9170
0.0400
0.0046
0.0019
0.0002
low_veg
0.0400
0.0249
0.8387
0.0961
0.0003
0.0000
tree
0.0249
0.0049
0.1133
0.8569
0.0000
0.0000
car
0.2946
0.0211
0.0408
0.0111
0.6324
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.898
0.940
0.844
0.842
0.800
--
Recall/Completeness
0.920
0.917
0.839
0.857
0.632
--
F1
0.909
0.929
0.841
0.849
0.706
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9459
0.0211
0.0269
0.0037
0.0024
0.0000
building
0.0286
0.9405
0.0274
0.0016
0.0017
0.0002
low_veg
0.0246
0.0195
0.8752
0.0804
0.0002
0.0000
tree
0.0090
0.0019
0.0918
0.8973
0.0000
0.0000
car
0.2012
0.0246
0.0275
0.0062
0.7405
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.935
0.955
0.883
0.878
0.828
--
Recall/Completeness
0.946
0.941
0.875
0.897
0.740
--
F1
0.941
0.948
0.879
0.887
0.782
--

Overall accuracy 0.914

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image