Detailled semantic labeling (2D) result


Name W. Sun
Affiliation Univ. of Calgary, Canada
Abbreviation UCal3
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
0.881
0.044
0.036
0.032
0.007
0.000
building
0.046
0.927
0.021
0.006
0.001
0.000
low_veg
0.047
0.038
0.692
0.222
0.001
0.000
tree
0.012
0.008
0.048
0.932
0.000
0.000
car
0.158
0.048
0.009
0.009
0.773
0.002
clutter
0.300
0.347
0.009
0.013
0.018
0.314
Precision/Correctness
0.893
0.903
0.850
0.783
0.798
0.949
Recall/Completeness
0.881
0.927
0.692
0.932
0.773
0.314
F1
0.887
0.915
0.763
0.851
0.786
0.472

Overall accuracy 0.857

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
0.916
0.033
0.029
0.017
0.004
0.000
building
0.042
0.938
0.017
0.003
0.000
0.000
low_veg
0.038
0.026
0.734
0.202
0.000
0.000
tree
0.005
0.003
0.039
0.953
0.000
0.000
car
0.085
0.051
0.006
0.005
0.851
0.002
clutter
0.265
0.366
0.006
0.006
0.017
0.340
Precision/Correctness
0.916
0.930
0.878
0.816
0.850
0.972
Recall/Completeness
0.916
0.938
0.734
0.953
0.851
0.340
F1
0.916
0.934
0.800
0.879
0.851
0.504

Overall accuracy 0.886
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9240
0.0203
0.0208
0.0275
0.0074
0.0000
building
0.0416
0.9350
0.0179
0.0053
0.0002
0.0000
low_veg
0.0621
0.0232
0.6343
0.2802
0.0002
0.0000
tree
0.0163
0.0054
0.0445
0.9332
0.0003
0.0004
car
0.1903
0.0257
0.0014
0.0035
0.7792
0.0000
clutter
0.0000
0.0000
0.0035
0.0297
0.0000
0.9668
Precision/Correctness
0.879
0.945
0.884
0.749
0.857
0.995
Recall/Completeness
0.924
0.935
0.634
0.933
0.779
0.967
F1
0.901
0.940
0.739
0.831
0.816
0.981

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9569
0.0146
0.0123
0.0113
0.0049
0.0000
building
0.0373
0.9487
0.0121
0.0017
0.0002
0.0000
low_veg
0.0488
0.0121
0.6694
0.2696
0.0001
0.0000
tree
0.0068
0.0016
0.0376
0.9538
0.0002
0.0001
car
0.0882
0.0226
0.0003
0.0005
0.8884
0.0000
clutter
0.0000
0.0000
0.0030
0.0113
0.0000
0.9856
Precision/Correctness
0.910
0.969
0.913
0.778
0.889
0.999
Recall/Completeness
0.957
0.949
0.669
0.954
0.888
0.986
F1
0.933
0.959
0.772
0.857
0.889
0.992

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9122
0.0599
0.0107
0.0100
0.0071
0.0000
building
0.0625
0.9252
0.0073
0.0035
0.0014
0.0000
low_veg
0.1132
0.0886
0.4110
0.3839
0.0033
0.0000
tree
0.0286
0.0112
0.0428
0.9170
0.0004
0.0000
car
0.2201
0.0156
0.0123
0.0065
0.7455
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.856
0.959
0.659
0.665
0.781
--
Recall/Completeness
0.912
0.925
0.411
0.917
0.746
--
F1
0.883
0.942
0.506
0.771
0.763
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9531
0.0304
0.0080
0.0049
0.0035
0.0000
building
0.0549
0.9352
0.0067
0.0018
0.0014
0.0000
low_veg
0.0850
0.0539
0.4448
0.4141
0.0022
0.0000
tree
0.0101
0.0032
0.0374
0.9493
0.0001
0.0000
car
0.1379
0.0101
0.0119
0.0039
0.8361
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.878
0.980
0.687
0.700
0.827
--
Recall/Completeness
0.953
0.935
0.445
0.949
0.836
--
F1
0.914
0.957
0.540
0.806
0.831
--

Overall accuracy 0.920

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
0.8759
0.0617
0.0198
0.0296
0.0130
0.0000
building
0.0569
0.9279
0.0088
0.0062
0.0002
0.0000
low_veg
0.0762
0.0456
0.6065
0.2705
0.0012
0.0000
tree
0.0079
0.0042
0.0295
0.9582
0.0001
0.0000
car
0.1848
0.0504
0.0025
0.0086
0.7537
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.874
0.931
0.797
0.877
0.776
--
Recall/Completeness
0.876
0.928
0.607
0.958
0.754
--
F1
0.875
0.930
0.689
0.916
0.765
--

Overall accuracy 0.886

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9247
0.0399
0.0139
0.0152
0.0063
0.0000
building
0.0520
0.9392
0.0062
0.0025
0.0001
0.0000
low_veg
0.0633
0.0238
0.6701
0.2421
0.0008
0.0000
tree
0.0024
0.0012
0.0235
0.9729
0.0001
0.0000
car
0.1380
0.0426
0.0010
0.0044
0.8140
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.897
0.961
0.838
0.911
0.854
--
Recall/Completeness
0.925
0.939
0.670
0.973
0.814
--
F1
0.911
0.950
0.745
0.941
0.834
--

Overall accuracy 0.918

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
0.8696
0.0400
0.0566
0.0270
0.0064
0.0004
building
0.0525
0.9294
0.0126
0.0052
0.0003
0.0000
low_veg
0.0422
0.0222
0.8075
0.1275
0.0006
0.0000
tree
0.0117
0.0075
0.0349
0.9457
0.0001
0.0001
car
0.1528
0.0246
0.0134
0.0097
0.7993
0.0002
clutter
--
--
--
--
--
--
Precision/Correctness
0.914
0.918
0.845
0.866
0.860
--
Recall/Completeness
0.870
0.929
0.807
0.946
0.799
--
F1
0.891
0.924
0.826
0.904
0.828
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9064
0.0300
0.0486
0.0120
0.0027
0.0004
building
0.0471
0.9431
0.0074
0.0021
0.0003
0.0000
low_veg
0.0295
0.0116
0.8499
0.1089
0.0002
0.0000
tree
0.0036
0.0030
0.0249
0.9684
0.0000
0.0001
car
0.0769
0.0209
0.0072
0.0068
0.8883
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.938
0.947
0.878
0.901
0.924
--
Recall/Completeness
0.906
0.943
0.850
0.968
0.888
--
F1
0.922
0.945
0.863
0.934
0.906
--

Overall accuracy 0.919

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
0.7978
0.1071
0.0507
0.0375
0.0070
0.0000
building
0.0339
0.9395
0.0149
0.0114
0.0002
0.0000
low_veg
0.0653
0.0448
0.6287
0.2612
0.0001
0.0000
tree
0.0073
0.0094
0.0434
0.9398
0.0000
0.0000
car
0.1098
0.0551
0.0275
0.0043
0.8033
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.932
0.815
0.777
0.755
0.678
--
Recall/Completeness
0.798
0.940
0.629
0.940
0.803
--
F1
0.860
0.873
0.695
0.837
0.736
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8352
0.0938
0.0451
0.0208
0.0051
0.0000
building
0.0315
0.9477
0.0123
0.0083
0.0002
0.0000
low_veg
0.0622
0.0279
0.6817
0.2281
0.0001
0.0000
tree
0.0028
0.0041
0.0343
0.9588
0.0000
0.0000
car
0.0456
0.0544
0.0250
0.0002
0.8747
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.944
0.845
0.808
0.798
0.694
--
Recall/Completeness
0.835
0.948
0.682
0.959
0.875
--
F1
0.886
0.893
0.740
0.871
0.774
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8516
0.0538
0.0399
0.0479
0.0052
0.0015
building
0.0968
0.8962
0.0032
0.0039
0.0000
0.0000
low_veg
0.0469
0.0511
0.6423
0.2590
0.0006
0.0000
tree
0.0065
0.0117
0.0518
0.9298
0.0000
0.0001
car
0.1115
0.0035
0.0045
0.0085
0.8720
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.775
0.865
0.880
0.806
0.826
--
Recall/Completeness
0.852
0.896
0.642
0.930
0.872
--
F1
0.812
0.880
0.743
0.864
0.848
--

Overall accuracy 0.828

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
0.8858
0.0484
0.0337
0.0285
0.0019
0.0016
building
0.1025
0.8956
0.0008
0.0011
0.0000
0.0000
low_veg
0.0463
0.0340
0.6884
0.2307
0.0005
0.0000
tree
0.0029
0.0045
0.0433
0.9492
0.0000
0.0001
car
0.0466
0.0027
0.0005
0.0008
0.9494
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.787
0.905
0.903
0.835
0.892
--
Recall/Completeness
0.886
0.896
0.688
0.949
0.949
--
F1
0.833
0.900
0.781
0.888
0.920
--

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
0.7898
0.0510
0.0951
0.0564
0.0077
0.0000
building
0.0187
0.9704
0.0061
0.0049
0.0000
0.0000
low_veg
0.0489
0.0872
0.6015
0.2621
0.0004
0.0000
tree
0.0149
0.0200
0.0639
0.9012
0.0000
0.0000
car
0.0790
0.0000
0.0000
0.0078
0.9131
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.896
0.808
0.775
0.790
0.567
--
Recall/Completeness
0.790
0.970
0.601
0.901
0.913
--
F1
0.840
0.882
0.677
0.842
0.699
--

Overall accuracy 0.811

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
0.8152
0.0439
0.0936
0.0406
0.0066
0.0000
building
0.0179
0.9777
0.0029
0.0015
0.0000
0.0000
low_veg
0.0456
0.0665
0.6382
0.2494
0.0003
0.0000
tree
0.0089
0.0115
0.0572
0.9224
0.0000
0.0000
car
0.0112
0.0000
0.0000
0.0031
0.9857
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.916
0.847
0.792
0.813
0.547
--
Recall/Completeness
0.815
0.978
0.638
0.922
0.986
--
F1
0.863
0.908
0.707
0.864
0.704
--

Overall accuracy 0.836

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
0.8072
0.0543
0.0538
0.0800
0.0048
0.0000
building
0.0387
0.9332
0.0183
0.0096
0.0001
0.0000
low_veg
0.0186
0.0366
0.7666
0.1781
0.0001
0.0000
tree
0.0158
0.0111
0.0647
0.9082
0.0001
0.0000
car
0.0632
0.0392
0.0087
0.0102
0.8786
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.872
0.876
0.866
0.833
0.701
--
Recall/Completeness
0.807
0.933
0.767
0.908
0.879
--
F1
0.838
0.904
0.813
0.869
0.780
--

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_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8480
0.0465
0.0493
0.0537
0.0026
0.0000
building
0.0381
0.9440
0.0137
0.0041
0.0000
0.0000
low_veg
0.0150
0.0267
0.7991
0.1591
0.0001
0.0000
tree
0.0100
0.0055
0.0559
0.9284
0.0001
0.0000
car
0.0192
0.0449
0.0050
0.0014
0.9296
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.895
0.908
0.886
0.859
0.766
--
Recall/Completeness
0.848
0.944
0.799
0.928
0.930
--
F1
0.871
0.926
0.840
0.892
0.840
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8428
0.0442
0.0373
0.0707
0.0050
0.0000
building
0.0077
0.9710
0.0149
0.0064
0.0000
0.0000
low_veg
0.0474
0.0512
0.6263
0.2747
0.0004
0.0000
tree
0.0040
0.0030
0.0215
0.9715
0.0000
0.0000
car
0.0521
0.0080
0.0122
0.0164
0.9113
0.0000
clutter
0.0167
0.9817
0.0000
0.0016
0.0000
0.0000
Precision/Correctness
0.930
0.773
0.866
0.853
0.778
0.000
Recall/Completeness
0.843
0.971
0.626
0.972
0.911
0.000
F1
0.884
0.861
0.727
0.909
0.839
-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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8844
0.0323
0.0364
0.0450
0.0020
0.0000
building
0.0068
0.9775
0.0140
0.0018
0.0000
0.0000
low_veg
0.0429
0.0374
0.6791
0.2402
0.0004
0.0000
tree
0.0013
0.0006
0.0159
0.9822
0.0000
0.0000
car
0.0053
0.0070
0.0018
0.0064
0.9795
0.0000
clutter
0.0138
0.9859
0.0000
0.0003
0.0000
0.0000
Precision/Correctness
0.947
0.797
0.885
0.884
0.862
-nan
Recall/Completeness
0.884
0.977
0.679
0.982
0.979
0.000
F1
0.914
0.878
0.769
0.931
0.917
-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
0.8649
0.0367
0.0353
0.0528
0.0103
0.0000
building
0.0201
0.9671
0.0051
0.0072
0.0006
0.0000
low_veg
0.0723
0.0541
0.6203
0.2526
0.0007
0.0000
tree
0.0118
0.0059
0.0235
0.9586
0.0002
0.0000
car
0.1479
0.0190
0.0044
0.0186
0.8101
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.923
0.875
0.868
0.799
0.713
--
Recall/Completeness
0.865
0.967
0.620
0.959
0.810
--
F1
0.893
0.919
0.724
0.872
0.758
--

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_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9063
0.0261
0.0310
0.0299
0.0066
0.0000
building
0.0178
0.9763
0.0020
0.0035
0.0005
0.0000
low_veg
0.0714
0.0349
0.6757
0.2175
0.0005
0.0000
tree
0.0041
0.0020
0.0156
0.9782
0.0000
0.0000
car
0.0843
0.0211
0.0021
0.0066
0.8859
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.939
0.917
0.898
0.842
0.755
--
Recall/Completeness
0.906
0.976
0.676
0.978
0.886
--
F1
0.922
0.946
0.771
0.905
0.815
--

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_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.7834
0.0556
0.0842
0.0632
0.0136
0.0000
building
0.0411
0.8991
0.0505
0.0093
0.0001
0.0000
low_veg
0.0216
0.0275
0.8196
0.1312
0.0001
0.0000
tree
0.0092
0.0109
0.0843
0.8954
0.0001
0.0000
car
0.0913
0.0149
0.0028
0.0088
0.8822
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.855
0.884
0.886
0.778
0.620
--
Recall/Completeness
0.783
0.899
0.820
0.895
0.882
--
F1
0.818
0.891
0.852
0.832
0.728
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8215
0.0432
0.0791
0.0459
0.0103
0.0000
building
0.0402
0.9054
0.0478
0.0066
0.0001
0.0000
low_veg
0.0193
0.0206
0.8464
0.1136
0.0001
0.0000
tree
0.0049
0.0061
0.0711
0.9178
0.0001
0.0000
car
0.0399
0.0134
0.0002
0.0039
0.9425
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.873
0.914
0.903
0.805
0.622
--
Recall/Completeness
0.822
0.905
0.846
0.918
0.943
--
F1
0.847
0.910
0.874
0.858
0.750
--

Overall accuracy 0.872

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
0.9215
0.0311
0.0208
0.0151
0.0115
0.0000
building
0.0769
0.8955
0.0200
0.0070
0.0006
0.0000
low_veg
0.0611
0.0373
0.5631
0.3379
0.0007
0.0000
tree
0.0048
0.0049
0.0125
0.9776
0.0002
0.0000
car
0.1978
0.0157
0.0076
0.0085
0.7703
0.0001
clutter
0.6209
0.2070
0.1077
0.0096
0.0547
0.0000
Precision/Correctness
0.862
0.943
0.877
0.571
0.857
0.000
Recall/Completeness
0.922
0.895
0.563
0.978
0.770
0.000
F1
0.891
0.919
0.686
0.721
0.812
-nan

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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9557
0.0189
0.0143
0.0072
0.0039
0.0000
building
0.0693
0.9107
0.0164
0.0032
0.0004
0.0000
low_veg
0.0474
0.0211
0.6042
0.3270
0.0003
0.0000
tree
0.0012
0.0014
0.0075
0.9899
0.0000
0.0000
car
0.1100
0.0143
0.0053
0.0062
0.8640
0.0001
clutter
0.6190
0.2322
0.0804
0.0040
0.0645
0.0000
Precision/Correctness
0.888
0.968
0.908
0.596
0.931
0.000
Recall/Completeness
0.956
0.911
0.604
0.990
0.864
0.000
F1
0.921
0.939
0.726
0.744
0.896
-nan

Overall accuracy 0.872

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
0.9281
0.0221
0.0189
0.0231
0.0075
0.0003
building
0.0249
0.9599
0.0063
0.0089
0.0001
0.0000
low_veg
0.0449
0.0295
0.5230
0.4019
0.0006
0.0000
tree
0.0016
0.0013
0.0047
0.9924
0.0000
0.0000
car
0.1625
0.0211
0.0036
0.0167
0.7710
0.0251
clutter
0.4002
0.4937
0.0242
0.0189
0.0279
0.0352
Precision/Correctness
0.934
0.900
0.876
0.600
0.716
0.689
Recall/Completeness
0.928
0.960
0.523
0.992
0.771
0.035
F1
0.931
0.929
0.655
0.748
0.742
0.067

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9549
0.0140
0.0129
0.0125
0.0054
0.0003
building
0.0204
0.9714
0.0033
0.0047
0.0001
0.0000
low_veg
0.0291
0.0140
0.5635
0.3931
0.0003
0.0000
tree
0.0004
0.0001
0.0023
0.9973
0.0000
0.0000
car
0.0869
0.0229
0.0023
0.0138
0.8442
0.0299
clutter
0.3757
0.5264
0.0184
0.0101
0.0286
0.0408
Precision/Correctness
0.952
0.924
0.914
0.635
0.731
0.721
Recall/Completeness
0.955
0.971
0.563
0.997
0.844
0.041
F1
0.953
0.947
0.697
0.776
0.784
0.077

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9557
0.0137
0.0127
0.0085
0.0090
0.0004
building
0.0490
0.9399
0.0080
0.0017
0.0014
0.0000
low_veg
0.0541
0.0043
0.8774
0.0625
0.0014
0.0003
tree
0.0231
0.0073
0.0522
0.9167
0.0007
0.0000
car
0.1027
0.1747
0.0063
0.0052
0.7110
0.0001
clutter
0.4340
0.3945
0.0068
0.0166
0.0434
0.1047
Precision/Correctness
0.897
0.908
0.906
0.883
0.820
0.940
Recall/Completeness
0.956
0.940
0.877
0.917
0.711
0.105
F1
0.926
0.924
0.892
0.899
0.761
0.188

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9741
0.0088
0.0103
0.0031
0.0035
0.0001
building
0.0401
0.9538
0.0046
0.0004
0.0010
0.0000
low_veg
0.0370
0.0008
0.9110
0.0508
0.0002
0.0001
tree
0.0069
0.0015
0.0402
0.9513
0.0001
0.0000
car
0.0485
0.1910
0.0025
0.0008
0.7572
0.0000
clutter
0.3930
0.4328
0.0051
0.0097
0.0439
0.1155
Precision/Correctness
0.924
0.926
0.932
0.918
0.878
0.979
Recall/Completeness
0.974
0.954
0.911
0.951
0.757
0.116
F1
0.948
0.939
0.921
0.934
0.813
0.207

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
0.9660
0.0137
0.0061
0.0077
0.0028
0.0037
building
0.0792
0.8070
0.1075
0.0057
0.0000
0.0005
low_veg
0.0560
0.0196
0.8669
0.0570
0.0005
0.0000
tree
0.0282
0.0063
0.0881
0.8773
0.0000
0.0000
car
0.1886
0.0443
0.0031
0.0055
0.7514
0.0071
clutter
0.4257
0.0789
0.0011
0.0021
0.0079
0.4842
Precision/Correctness
0.861
0.944
0.757
0.871
0.858
0.940
Recall/Completeness
0.966
0.807
0.867
0.877
0.751
0.484
F1
0.911
0.870
0.808
0.874
0.801
0.639

Overall accuracy 0.867

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9839
0.0084
0.0021
0.0032
0.0014
0.0010
building
0.0754
0.8126
0.1088
0.0028
0.0000
0.0004
low_veg
0.0283
0.0107
0.9110
0.0497
0.0002
0.0000
tree
0.0106
0.0016
0.0825
0.9053
0.0000
0.0000
car
0.0991
0.0532
0.0007
0.0027
0.8379
0.0064
clutter
0.3957
0.0788
0.0004
0.0009
0.0079
0.5163
Precision/Correctness
0.886
0.963
0.766
0.906
0.897
0.979
Recall/Completeness
0.984
0.813
0.911
0.905
0.838
0.516
F1
0.932
0.881
0.833
0.906
0.866
0.676

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8122
0.0587
0.0834
0.0432
0.0026
0.0000
building
0.0212
0.9563
0.0149
0.0073
0.0004
0.0000
low_veg
0.0272
0.0167
0.6259
0.3296
0.0005
0.0000
tree
0.0178
0.0062
0.0491
0.9267
0.0001
0.0000
car
0.1625
0.0233
0.0000
0.0257
0.7885
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.920
0.926
0.810
0.696
0.868
--
Recall/Completeness
0.812
0.956
0.626
0.927
0.788
--
F1
0.863
0.941
0.706
0.795
0.826
--

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_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8698
0.0492
0.0635
0.0164
0.0011
0.0000
building
0.0174
0.9735
0.0068
0.0020
0.0003
0.0000
low_veg
0.0192
0.0060
0.6569
0.3174
0.0005
0.0000
tree
0.0056
0.0015
0.0344
0.9586
0.0000
0.0000
car
0.0788
0.0246
0.0000
0.0157
0.8809
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.950
0.951
0.864
0.727
0.914
--
Recall/Completeness
0.870
0.974
0.657
0.959
0.881
--
F1
0.908
0.962
0.746
0.827
0.897
--

Overall accuracy 0.867

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8801
0.0498
0.0373
0.0273
0.0053
0.0002
building
0.0167
0.9402
0.0377
0.0040
0.0013
0.0000
low_veg
0.0441
0.0435
0.7698
0.1421
0.0005
0.0000
tree
0.0133
0.0063
0.0734
0.9069
0.0001
0.0000
car
0.1486
0.0246
0.0345
0.0107
0.7816
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.920
0.898
0.863
0.786
0.795
--
Recall/Completeness
0.880
0.940
0.770
0.907
0.782
--
F1
0.899
0.919
0.814
0.842
0.788
--

Overall accuracy 0.869

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
0.9160
0.0442
0.0251
0.0121
0.0024
0.0002
building
0.0136
0.9540
0.0296
0.0016
0.0012
0.0000
low_veg
0.0339
0.0355
0.8101
0.1202
0.0003
0.0000
tree
0.0045
0.0016
0.0593
0.9346
0.0000
0.0000
car
0.0788
0.0239
0.0236
0.0056
0.8681
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.945
0.919
0.896
0.828
0.861
--
Recall/Completeness
0.916
0.954
0.810
0.935
0.868
--
F1
0.930
0.936
0.851
0.878
0.864
--

Overall accuracy 0.900

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image