Detailled semantic labeling (2D) result


Name W. Tang
Affiliation Wuhan University, China
Abbreviation WUH_W5
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
0.912
0.029
0.038
0.016
0.005
0.000
building
0.039
0.933
0.022
0.005
0.000
0.000
low_veg
0.056
0.021
0.786
0.137
0.000
0.000
tree
0.022
0.005
0.097
0.875
0.000
0.000
car
0.244
0.028
0.006
0.003
0.718
0.000
clutter
0.336
0.291
0.018
0.007
0.022
0.326
Precision/Correctness
0.883
0.936
0.815
0.848
0.848
0.942
Recall/Completeness
0.912
0.933
0.786
0.875
0.718
0.326
F1
0.897
0.934
0.800
0.861
0.777
0.484

Overall accuracy 0.874

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
0.937
0.022
0.030
0.008
0.003
0.000
building
0.031
0.952
0.015
0.002
0.000
0.000
low_veg
0.043
0.013
0.821
0.123
0.000
0.000
tree
0.011
0.002
0.079
0.908
0.000
0.000
car
0.132
0.032
0.003
0.001
0.830
0.001
clutter
0.299
0.307
0.015
0.002
0.022
0.355
Precision/Correctness
0.917
0.954
0.851
0.877
0.888
0.958
Recall/Completeness
0.937
0.952
0.821
0.908
0.830
0.355
F1
0.927
0.953
0.836
0.892
0.858
0.518

Overall accuracy 0.904
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9462
0.0162
0.0197
0.0119
0.0060
0.0000
building
0.0369
0.9386
0.0209
0.0037
0.0000
0.0000
low_veg
0.0714
0.0186
0.7338
0.1761
0.0000
0.0000
tree
0.0304
0.0049
0.0899
0.8724
0.0005
0.0019
car
0.2760
0.0012
0.0003
0.0008
0.7217
0.0000
clutter
0.0000
0.0000
0.0000
0.0113
0.0000
0.9887
Precision/Correctness
0.863
0.956
0.849
0.821
0.873
0.975
Recall/Completeness
0.946
0.939
0.734
0.872
0.722
0.989
F1
0.902
0.947
0.787
0.846
0.790
0.982

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9689
0.0126
0.0106
0.0038
0.0041
0.0000
building
0.0281
0.9586
0.0127
0.0006
0.0000
0.0000
low_veg
0.0522
0.0107
0.7665
0.1706
0.0000
0.0000
tree
0.0144
0.0021
0.0750
0.9072
0.0003
0.0010
car
0.1418
0.0002
0.0000
0.0002
0.8578
0.0000
clutter
0.0000
0.0000
0.0000
0.0024
0.0000
0.9975
Precision/Correctness
0.907
0.973
0.883
0.843
0.902
0.988
Recall/Completeness
0.969
0.959
0.766
0.907
0.858
0.998
F1
0.937
0.966
0.821
0.874
0.879
0.993

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9336
0.0460
0.0097
0.0066
0.0041
0.0000
building
0.0634
0.9247
0.0094
0.0017
0.0008
0.0000
low_veg
0.1332
0.0541
0.4901
0.3213
0.0013
0.0000
tree
0.0407
0.0172
0.0560
0.8860
0.0001
0.0000
car
0.2951
0.0145
0.0004
0.0017
0.6883
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.851
0.969
0.669
0.711
0.856
--
Recall/Completeness
0.934
0.925
0.490
0.886
0.688
--
F1
0.890
0.946
0.566
0.789
0.763
--

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
0.9606
0.0278
0.0073
0.0024
0.0019
0.0000
building
0.0502
0.9404
0.0082
0.0005
0.0007
0.0000
low_veg
0.0945
0.0267
0.5310
0.3468
0.0011
0.0000
tree
0.0170
0.0074
0.0494
0.9262
0.0000
0.0000
car
0.1756
0.0120
0.0001
0.0003
0.8120
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.884
0.983
0.698
0.745
0.899
--
Recall/Completeness
0.961
0.940
0.531
0.926
0.812
--
F1
0.921
0.961
0.603
0.826
0.853
--

Overall accuracy 0.927

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
0.9089
0.0455
0.0316
0.0095
0.0045
0.0000
building
0.0332
0.9486
0.0140
0.0040
0.0002
0.0000
low_veg
0.0927
0.0281
0.7268
0.1521
0.0002
0.0000
tree
0.0193
0.0033
0.0645
0.9127
0.0002
0.0000
car
0.3230
0.0111
0.0004
0.0014
0.6640
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.880
0.952
0.713
0.929
0.894
--
Recall/Completeness
0.909
0.949
0.727
0.913
0.664
--
F1
0.894
0.950
0.720
0.921
0.762
--

Overall accuracy 0.900

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9409
0.0313
0.0232
0.0028
0.0017
0.0000
building
0.0237
0.9656
0.0093
0.0013
0.0000
0.0000
low_veg
0.0756
0.0125
0.7773
0.1345
0.0000
0.0000
tree
0.0080
0.0010
0.0518
0.9390
0.0001
0.0000
car
0.2121
0.0039
0.0001
0.0000
0.7839
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.920
0.972
0.756
0.951
0.953
--
Recall/Completeness
0.941
0.966
0.777
0.939
0.784
--
F1
0.930
0.969
0.767
0.945
0.860
--

Overall accuracy 0.931

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
0.8919
0.0237
0.0703
0.0098
0.0042
0.0000
building
0.0351
0.9422
0.0179
0.0047
0.0001
0.0000
low_veg
0.0578
0.0114
0.8855
0.0450
0.0003
0.0000
tree
0.0277
0.0058
0.0881
0.8784
0.0001
0.0000
car
0.2642
0.0039
0.0132
0.0021
0.7166
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.902
0.951
0.786
0.941
0.896
--
Recall/Completeness
0.892
0.942
0.885
0.878
0.717
--
F1
0.897
0.947
0.833
0.909
0.796
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9168
0.0174
0.0604
0.0031
0.0022
0.0000
building
0.0247
0.9632
0.0102
0.0019
0.0000
0.0000
low_veg
0.0410
0.0038
0.9212
0.0340
0.0000
0.0000
tree
0.0127
0.0023
0.0666
0.9184
0.0000
0.0000
car
0.1379
0.0015
0.0079
0.0011
0.8517
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.939
0.972
0.827
0.965
0.938
--
Recall/Completeness
0.917
0.963
0.921
0.918
0.852
--
F1
0.928
0.967
0.872
0.941
0.893
--

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_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8811
0.0549
0.0427
0.0167
0.0041
0.0005
building
0.0336
0.9443
0.0114
0.0095
0.0012
0.0000
low_veg
0.0887
0.0259
0.7334
0.1517
0.0003
0.0000
tree
0.0195
0.0086
0.0851
0.8866
0.0001
0.0000
car
0.2273
0.0453
0.0019
0.0002
0.7252
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.919
0.891
0.781
0.838
0.727
--
Recall/Completeness
0.881
0.944
0.733
0.887
0.725
--
F1
0.900
0.917
0.756
0.862
0.726
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9098
0.0444
0.0353
0.0070
0.0030
0.0005
building
0.0263
0.9589
0.0073
0.0064
0.0010
0.0000
low_veg
0.0842
0.0141
0.7746
0.1268
0.0003
0.0000
tree
0.0085
0.0044
0.0650
0.9220
0.0000
0.0000
car
0.1297
0.0485
0.0008
0.0000
0.8210
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.938
0.918
0.822
0.878
0.748
--
Recall/Completeness
0.910
0.959
0.775
0.922
0.821
--
F1
0.924
0.938
0.798
0.899
0.783
--

Overall accuracy 0.901

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
0.8822
0.0381
0.0495
0.0284
0.0017
0.0000
building
0.0251
0.9515
0.0167
0.0066
0.0000
0.0000
low_veg
0.0387
0.0191
0.7662
0.1757
0.0001
0.0001
tree
0.0088
0.0053
0.1074
0.8785
0.0000
0.0000
car
0.2262
0.0003
0.0002
0.0059
0.7675
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.860
0.934
0.816
0.852
0.933
--
Recall/Completeness
0.882
0.952
0.766
0.878
0.767
--
F1
0.871
0.942
0.791
0.865
0.842
--

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_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9068
0.0365
0.0422
0.0143
0.0002
0.0000
building
0.0235
0.9686
0.0060
0.0019
0.0000
0.0000
low_veg
0.0345
0.0120
0.7983
0.1551
0.0000
0.0001
tree
0.0037
0.0017
0.0888
0.9058
0.0000
0.0000
car
0.1039
0.0000
0.0000
0.0023
0.8938
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.889
0.953
0.849
0.879
0.990
--
Recall/Completeness
0.907
0.969
0.798
0.906
0.894
--
F1
0.898
0.961
0.823
0.892
0.940
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8520
0.0347
0.0752
0.0316
0.0051
0.0014
building
0.0093
0.9676
0.0175
0.0056
0.0000
0.0000
low_veg
0.0886
0.0501
0.7106
0.1503
0.0003
0.0000
tree
0.0252
0.0103
0.1331
0.8314
0.0001
0.0000
car
0.1568
0.0000
0.0000
0.0032
0.8400
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.855
0.876
0.736
0.857
0.639
--
Recall/Completeness
0.852
0.968
0.711
0.831
0.840
--
F1
0.854
0.919
0.723
0.844
0.726
--

Overall accuracy 0.829

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
0.8719
0.0304
0.0703
0.0217
0.0042
0.0015
building
0.0049
0.9849
0.0082
0.0020
0.0000
0.0000
low_veg
0.0828
0.0362
0.7397
0.1410
0.0003
0.0000
tree
0.0169
0.0045
0.1200
0.8586
0.0000
0.0000
car
0.0525
0.0000
0.0000
0.0003
0.9471
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.882
0.907
0.759
0.878
0.637
--
Recall/Completeness
0.872
0.985
0.740
0.859
0.947
--
F1
0.877
0.944
0.749
0.868
0.762
--

Overall accuracy 0.854

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
0.8352
0.0221
0.0898
0.0510
0.0018
0.0000
building
0.0222
0.9366
0.0319
0.0090
0.0001
0.0001
low_veg
0.0139
0.0125
0.8459
0.1268
0.0000
0.0009
tree
0.0226
0.0032
0.1156
0.8586
0.0000
0.0000
car
0.1826
0.0009
0.0178
0.0010
0.7977
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.885
0.953
0.802
0.870
0.855
--
Recall/Completeness
0.835
0.937
0.846
0.859
0.798
--
F1
0.859
0.945
0.823
0.864
0.826
--

Overall accuracy 0.865

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area16_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8697
0.0172
0.0810
0.0314
0.0007
0.0000
building
0.0185
0.9626
0.0158
0.0029
0.0000
0.0002
low_veg
0.0100
0.0077
0.8681
0.1132
0.0000
0.0010
tree
0.0152
0.0009
0.1002
0.8837
0.0000
0.0000
car
0.0966
0.0000
0.0048
0.0001
0.8985
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.916
0.971
0.832
0.893
0.934
--
Recall/Completeness
0.870
0.963
0.868
0.884
0.898
--
F1
0.892
0.967
0.850
0.888
0.916
--

Overall accuracy 0.891

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8743
0.0332
0.0378
0.0491
0.0055
0.0000
building
0.0118
0.9700
0.0108
0.0075
0.0000
0.0000
low_veg
0.0544
0.0343
0.7317
0.1796
0.0000
0.0000
tree
0.0086
0.0022
0.0437
0.9455
0.0000
0.0000
car
0.1109
0.0017
0.0240
0.0058
0.8577
0.0000
clutter
0.0031
0.9954
0.0007
0.0009
0.0000
0.0000
Precision/Correctness
0.914
0.794
0.838
0.894
0.759
-nan
Recall/Completeness
0.874
0.970
0.732
0.945
0.858
0.000
F1
0.894
0.873
0.781
0.919
0.805
-nan

Overall accuracy 0.868

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9080
0.0228
0.0355
0.0297
0.0039
0.0000
building
0.0086
0.9831
0.0062
0.0021
0.0000
0.0000
low_veg
0.0478
0.0239
0.7790
0.1493
0.0000
0.0000
tree
0.0035
0.0007
0.0325
0.9633
0.0000
0.0000
car
0.0259
0.0000
0.0118
0.0000
0.9624
0.0000
clutter
0.0010
0.9987
0.0000
0.0003
0.0000
0.0000
Precision/Correctness
0.939
0.814
0.869
0.922
0.785
-nan
Recall/Completeness
0.908
0.983
0.779
0.963
0.962
0.000
F1
0.923
0.890
0.821
0.942
0.865
-nan

Overall accuracy 0.894

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9041
0.0200
0.0411
0.0296
0.0052
0.0000
building
0.0323
0.9420
0.0187
0.0070
0.0000
0.0000
low_veg
0.0873
0.0262
0.7546
0.1318
0.0001
0.0000
tree
0.0281
0.0036
0.0640
0.9039
0.0003
0.0000
car
0.2482
0.0002
0.0073
0.0105
0.7292
0.0046
clutter
--
--
--
--
--
--
Precision/Correctness
0.894
0.930
0.813
0.874
0.819
--
Recall/Completeness
0.904
0.942
0.755
0.904
0.729
--
F1
0.899
0.936
0.783
0.889
0.771
--

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_area22_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9328
0.0144
0.0341
0.0153
0.0032
0.0000
building
0.0246
0.9640
0.0087
0.0027
0.0000
0.0000
low_veg
0.0820
0.0168
0.7959
0.1053
0.0000
0.0000
tree
0.0132
0.0010
0.0463
0.9394
0.0001
0.0000
car
0.1416
0.0002
0.0037
0.0031
0.8444
0.0070
clutter
--
--
--
--
--
--
Precision/Correctness
0.923
0.955
0.857
0.912
0.864
--
Recall/Completeness
0.933
0.964
0.796
0.939
0.844
--
F1
0.928
0.959
0.825
0.925
0.854
--

Overall accuracy 0.913

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
0.8249
0.0418
0.0924
0.0344
0.0066
0.0000
building
0.0516
0.8837
0.0567
0.0079
0.0000
0.0000
low_veg
0.0236
0.0202
0.8815
0.0744
0.0003
0.0000
tree
0.0204
0.0064
0.1626
0.8103
0.0002
0.0000
car
0.2075
0.0236
0.0045
0.0051
0.7592
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.826
0.912
0.847
0.847
0.718
--
Recall/Completeness
0.825
0.884
0.882
0.810
0.759
--
F1
0.825
0.897
0.864
0.828
0.738
--

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
0.8539
0.0330
0.0826
0.0253
0.0051
0.0000
building
0.0487
0.8974
0.0488
0.0051
0.0000
0.0000
low_veg
0.0200
0.0161
0.8995
0.0641
0.0003
0.0000
tree
0.0134
0.0034
0.1399
0.8431
0.0001
0.0000
car
0.1024
0.0317
0.0001
0.0016
0.8642
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.852
0.933
0.870
0.870
0.725
--
Recall/Completeness
0.854
0.897
0.900
0.843
0.864
--
F1
0.853
0.915
0.885
0.856
0.789
--

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
0.9422
0.0252
0.0191
0.0067
0.0069
0.0000
building
0.0786
0.9003
0.0171
0.0037
0.0003
0.0000
low_veg
0.0765
0.0297
0.6964
0.1965
0.0008
0.0000
tree
0.0147
0.0055
0.0687
0.9109
0.0003
0.0000
car
0.2679
0.0084
0.0041
0.0027
0.7169
0.0000
clutter
0.4704
0.2629
0.1837
0.0097
0.0733
0.0000
Precision/Correctness
0.848
0.954
0.875
0.685
0.897
-nan
Recall/Completeness
0.942
0.900
0.696
0.911
0.717
0.000
F1
0.893
0.926
0.775
0.782
0.797
-nan

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_area27_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9665
0.0161
0.0127
0.0024
0.0023
0.0000
building
0.0675
0.9194
0.0117
0.0013
0.0002
0.0000
low_veg
0.0567
0.0171
0.7376
0.1879
0.0007
0.0000
tree
0.0069
0.0013
0.0500
0.9418
0.0000
0.0000
car
0.1462
0.0064
0.0027
0.0020
0.8427
0.0000
clutter
0.4814
0.2624
0.1671
0.0021
0.0869
0.0000
Precision/Correctness
0.884
0.973
0.911
0.714
0.949
-nan
Recall/Completeness
0.966
0.919
0.738
0.942
0.843
0.000
F1
0.923
0.946
0.815
0.812
0.893
-nan

Overall accuracy 0.898

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9424
0.0158
0.0240
0.0122
0.0055
0.0001
building
0.0273
0.9511
0.0149
0.0067
0.0000
0.0000
low_veg
0.0652
0.0159
0.6658
0.2528
0.0003
0.0000
tree
0.0060
0.0020
0.0305
0.9614
0.0000
0.0000
car
0.2187
0.0025
0.0008
0.0067
0.7712
0.0000
clutter
0.4251
0.4277
0.0378
0.0128
0.0432
0.0534
Precision/Correctness
0.923
0.922
0.841
0.701
0.739
0.974
Recall/Completeness
0.942
0.951
0.666
0.961
0.771
0.053
F1
0.933
0.936
0.743
0.811
0.755
0.101

Overall accuracy 0.877

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
0.9636
0.0099
0.0173
0.0053
0.0039
0.0000
building
0.0197
0.9683
0.0087
0.0033
0.0000
0.0000
low_veg
0.0419
0.0061
0.7118
0.2401
0.0001
0.0000
tree
0.0017
0.0003
0.0211
0.9769
0.0000
0.0000
car
0.1286
0.0002
0.0004
0.0038
0.8670
0.0000
clutter
0.4000
0.4578
0.0313
0.0053
0.0449
0.0607
Precision/Correctness
0.946
0.940
0.882
0.741
0.753
0.995
Recall/Completeness
0.964
0.968
0.712
0.977
0.867
0.061
F1
0.955
0.954
0.788
0.843
0.806
0.114

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9728
0.0094
0.0071
0.0046
0.0058
0.0003
building
0.0342
0.9560
0.0069
0.0021
0.0005
0.0002
low_veg
0.0488
0.0051
0.9076
0.0375
0.0009
0.0002
tree
0.0434
0.0042
0.0868
0.8649
0.0006
0.0000
car
0.1924
0.1246
0.0037
0.0030
0.6761
0.0002
clutter
0.5619
0.1792
0.0320
0.0100
0.0298
0.1871
Precision/Correctness
0.888
0.945
0.896
0.921
0.871
0.968
Recall/Completeness
0.973
0.956
0.908
0.865
0.676
0.187
F1
0.929
0.950
0.902
0.892
0.761
0.314

Overall accuracy 0.909

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
0.9853
0.0071
0.0041
0.0011
0.0024
0.0001
building
0.0226
0.9744
0.0020
0.0004
0.0003
0.0002
low_veg
0.0282
0.0029
0.9378
0.0308
0.0003
0.0000
tree
0.0176
0.0009
0.0687
0.9127
0.0001
0.0000
car
0.0968
0.1484
0.0009
0.0009
0.7528
0.0003
clutter
0.5251
0.1952
0.0314
0.0039
0.0310
0.2134
Precision/Correctness
0.925
0.955
0.929
0.950
0.915
0.984
Recall/Completeness
0.985
0.974
0.938
0.913
0.753
0.213
F1
0.954
0.964
0.934
0.931
0.826
0.351

Overall accuracy 0.937

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
0.9690
0.0151
0.0068
0.0045
0.0028
0.0017
building
0.0498
0.8982
0.0459
0.0041
0.0007
0.0014
low_veg
0.0761
0.0103
0.8820
0.0309
0.0007
0.0000
tree
0.0332
0.0059
0.1371
0.8236
0.0002
0.0000
car
0.1802
0.0251
0.0019
0.0006
0.7869
0.0054
clutter
0.4608
0.0784
0.0005
0.0012
0.0221
0.4370
Precision/Correctness
0.868
0.954
0.817
0.917
0.795
0.954
Recall/Completeness
0.969
0.898
0.882
0.824
0.787
0.437
F1
0.916
0.925
0.848
0.868
0.791
0.599

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_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9821
0.0119
0.0027
0.0015
0.0016
0.0003
building
0.0419
0.9106
0.0437
0.0020
0.0005
0.0012
low_veg
0.0438
0.0033
0.9247
0.0278
0.0004
0.0000
tree
0.0120
0.0018
0.1246
0.8615
0.0001
0.0000
car
0.0694
0.0189
0.0001
0.0001
0.9058
0.0057
clutter
0.4235
0.0748
0.0002
0.0008
0.0241
0.4767
Precision/Correctness
0.899
0.968
0.838
0.942
0.815
0.978
Recall/Completeness
0.982
0.911
0.925
0.862
0.906
0.477
F1
0.939
0.938
0.879
0.900
0.858
0.641

Overall accuracy 0.913

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
0.8738
0.0458
0.0618
0.0176
0.0011
0.0000
building
0.0315
0.9375
0.0238
0.0071
0.0001
0.0000
low_veg
0.0523
0.0126
0.7093
0.2258
0.0000
0.0000
tree
0.0384
0.0067
0.1059
0.8489
0.0001
0.0000
car
0.2754
0.0047
0.0016
0.0063
0.7120
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.871
0.939
0.790
0.761
0.943
--
Recall/Completeness
0.874
0.938
0.709
0.849
0.712
--
F1
0.872
0.938
0.748
0.803
0.811
--

Overall accuracy 0.843

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
0.9113
0.0409
0.0423
0.0053
0.0003
0.0000
building
0.0215
0.9636
0.0132
0.0016
0.0001
0.0000
low_veg
0.0357
0.0057
0.7396
0.2191
0.0000
0.0000
tree
0.0146
0.0022
0.0808
0.9023
0.0000
0.0000
car
0.1510
0.0062
0.0008
0.0021
0.8399
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.922
0.958
0.847
0.788
0.981
--
Recall/Completeness
0.911
0.964
0.740
0.902
0.840
--
F1
0.917
0.961
0.790
0.841
0.905
--

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8983
0.0292
0.0530
0.0152
0.0033
0.0009
building
0.0219
0.9249
0.0465
0.0055
0.0012
0.0000
low_veg
0.0401
0.0139
0.8591
0.0865
0.0004
0.0000
tree
0.0214
0.0044
0.1313
0.8427
0.0002
0.0000
car
0.2354
0.0231
0.0302
0.0046
0.7068
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.912
0.948
0.823
0.847
0.836
--
Recall/Completeness
0.898
0.925
0.859
0.843
0.707
--
F1
0.905
0.936
0.840
0.845
0.766
--

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
0.9254
0.0274
0.0388
0.0057
0.0016
0.0010
building
0.0143
0.9523
0.0303
0.0022
0.0009
0.0000
low_veg
0.0245
0.0099
0.8937
0.0717
0.0001
0.0000
tree
0.0077
0.0015
0.1060
0.8847
0.0000
0.0000
car
0.1324
0.0270
0.0190
0.0010
0.8206
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.950
0.960
0.865
0.884
0.892
--
Recall/Completeness
0.925
0.952
0.894
0.885
0.821
--
F1
0.938
0.956
0.879
0.885
0.855
--

Overall accuracy 0.915

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image