Detailled semantic labeling (2D) result


Name L. Pibre and J. Pasquet
Affiliation LIRMM Montpellier, France
Abbreviation LIRM_1
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
0.907
0.031
0.040
0.017
0.004
0.000
building
0.065
0.906
0.021
0.008
0.001
0.000
low_veg
0.066
0.023
0.771
0.140
0.000
0.000
tree
0.026
0.006
0.096
0.871
0.000
0.000
car
0.447
0.045
0.012
0.006
0.489
0.002
clutter
0.547
0.110
0.044
0.007
0.067
0.225
Precision/Correctness
0.841
0.934
0.811
0.842
0.757
0.944
Recall/Completeness
0.907
0.906
0.771
0.871
0.489
0.225
F1
0.873
0.920
0.790
0.856
0.594
0.363

Overall accuracy 0.858

Overall statistics, reference set: no_boundary


Number of non processed tiles by this participant: 0
↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.933
0.024
0.032
0.009
0.002
0.000
building
0.058
0.923
0.014
0.005
0.001
0.000
low_veg
0.053
0.014
0.809
0.124
0.000
0.000
tree
0.014
0.003
0.080
0.903
0.000
0.000
car
0.374
0.052
0.009
0.003
0.560
0.002
clutter
0.527
0.111
0.043
0.002
0.069
0.248
Precision/Correctness
0.872
0.955
0.845
0.871
0.787
0.960
Recall/Completeness
0.933
0.923
0.809
0.903
0.560
0.248
F1
0.901
0.939
0.827
0.887
0.654
0.395

Overall accuracy 0.888
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9493
0.0092
0.0249
0.0124
0.0042
0.0000
building
0.0341
0.9402
0.0201
0.0053
0.0003
0.0000
low_veg
0.1178
0.0222
0.6861
0.1737
0.0003
0.0000
tree
0.0427
0.0067
0.0846
0.8645
0.0002
0.0012
car
0.4661
0.0066
0.0009
0.0035
0.5216
0.0013
clutter
0.0260
0.0003
0.0000
0.0163
0.0000
0.9573
Precision/Correctness
0.812
0.958
0.841
0.819
0.869
0.982
Recall/Completeness
0.949
0.940
0.686
0.865
0.522
0.957
F1
0.875
0.949
0.756
0.841
0.652
0.969

Overall accuracy 0.856

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
0.9728
0.0053
0.0153
0.0040
0.0026
0.0000
building
0.0246
0.9581
0.0144
0.0025
0.0003
0.0000
low_veg
0.1007
0.0109
0.7208
0.1674
0.0002
0.0000
tree
0.0257
0.0031
0.0718
0.8989
0.0001
0.0005
car
0.3750
0.0062
0.0010
0.0016
0.6146
0.0016
clutter
0.0269
0.0003
0.0000
0.0036
0.0000
0.9692
Precision/Correctness
0.853
0.979
0.873
0.843
0.906
0.992
Recall/Completeness
0.973
0.958
0.721
0.899
0.615
0.969
F1
0.909
0.969
0.790
0.870
0.732
0.981

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_area4_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9087
0.0685
0.0126
0.0048
0.0054
0.0000
building
0.0515
0.9364
0.0101
0.0018
0.0001
0.0000
low_veg
0.1625
0.0632
0.5162
0.2580
0.0000
0.0000
tree
0.0532
0.0212
0.0759
0.8493
0.0005
0.0000
car
0.5493
0.0346
0.0021
0.0014
0.4113
0.0013
clutter
--
--
--
--
--
--
Precision/Correctness
0.850
0.956
0.638
0.745
0.779
--
Recall/Completeness
0.909
0.936
0.516
0.849
0.411
--
F1
0.879
0.946
0.571
0.794
0.538
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9394
0.0450
0.0096
0.0021
0.0039
0.0000
building
0.0396
0.9509
0.0085
0.0008
0.0001
0.0000
low_veg
0.1244
0.0295
0.5648
0.2814
0.0000
0.0000
tree
0.0256
0.0101
0.0715
0.8928
0.0000
0.0000
car
0.4936
0.0308
0.0018
0.0009
0.4714
0.0016
clutter
--
--
--
--
--
--
Precision/Correctness
0.885
0.974
0.670
0.772
0.815
--
Recall/Completeness
0.939
0.951
0.565
0.893
0.471
--
F1
0.911
0.962
0.613
0.828
0.597
--

Overall accuracy 0.923

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
0.9017
0.0538
0.0288
0.0088
0.0069
0.0001
building
0.0422
0.9351
0.0172
0.0049
0.0006
0.0000
low_veg
0.0981
0.0290
0.6922
0.1801
0.0005
0.0001
tree
0.0214
0.0039
0.0602
0.9143
0.0002
0.0000
car
0.4849
0.0237
0.0079
0.0028
0.4787
0.0019
clutter
--
--
--
--
--
--
Precision/Correctness
0.856
0.944
0.708
0.919
0.792
--
Recall/Completeness
0.902
0.935
0.692
0.914
0.479
--
F1
0.878
0.940
0.700
0.917
0.597
--

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_area6_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9356
0.0372
0.0199
0.0026
0.0047
0.0000
building
0.0326
0.9527
0.0124
0.0018
0.0005
0.0000
low_veg
0.0792
0.0161
0.7431
0.1614
0.0001
0.0000
tree
0.0091
0.0012
0.0494
0.9403
0.0001
0.0000
car
0.4105
0.0188
0.0089
0.0014
0.5582
0.0022
clutter
--
--
--
--
--
--
Precision/Correctness
0.897
0.966
0.751
0.943
0.834
--
Recall/Completeness
0.936
0.953
0.743
0.940
0.558
--
F1
0.916
0.959
0.747
0.942
0.669
--

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_area8_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8919
0.0275
0.0654
0.0107
0.0041
0.0004
building
0.0310
0.9481
0.0161
0.0039
0.0005
0.0002
low_veg
0.0644
0.0182
0.8584
0.0587
0.0003
0.0000
tree
0.0321
0.0075
0.0835
0.8759
0.0001
0.0009
car
0.4515
0.0150
0.0317
0.0046
0.4968
0.0005
clutter
--
--
--
--
--
--
Precision/Correctness
0.888
0.939
0.790
0.930
0.851
--
Recall/Completeness
0.892
0.948
0.858
0.876
0.497
--
F1
0.890
0.943
0.823
0.902
0.627
--

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
0.9175
0.0206
0.0559
0.0035
0.0021
0.0004
building
0.0219
0.9669
0.0088
0.0017
0.0004
0.0003
low_veg
0.0473
0.0096
0.8964
0.0467
0.0000
0.0000
tree
0.0151
0.0037
0.0642
0.9160
0.0000
0.0009
car
0.3806
0.0169
0.0247
0.0026
0.5746
0.0007
clutter
--
--
--
--
--
--
Precision/Correctness
0.926
0.961
0.831
0.954
0.905
--
Recall/Completeness
0.918
0.967
0.896
0.916
0.575
--
F1
0.922
0.964
0.863
0.935
0.703
--

Overall accuracy 0.921

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8514
0.0680
0.0571
0.0207
0.0028
0.0000
building
0.0225
0.9453
0.0208
0.0110
0.0004
0.0000
low_veg
0.0762
0.0289
0.7391
0.1556
0.0002
0.0000
tree
0.0176
0.0083
0.0817
0.8923
0.0000
0.0000
car
0.4528
0.0345
0.0095
0.0030
0.4997
0.0005
clutter
--
--
--
--
--
--
Precision/Correctness
0.926
0.873
0.751
0.829
0.748
--
Recall/Completeness
0.851
0.945
0.739
0.892
0.500
--
F1
0.887
0.908
0.745
0.860
0.599
--

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_area10_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8859
0.0546
0.0485
0.0094
0.0016
0.0000
building
0.0166
0.9598
0.0154
0.0079
0.0003
0.0000
low_veg
0.0722
0.0144
0.7865
0.1267
0.0003
0.0000
tree
0.0082
0.0041
0.0641
0.9236
0.0000
0.0000
car
0.3945
0.0402
0.0080
0.0010
0.5555
0.0007
clutter
--
--
--
--
--
--
Precision/Correctness
0.945
0.905
0.790
0.873
0.798
--
Recall/Completeness
0.886
0.960
0.787
0.924
0.555
--
F1
0.914
0.931
0.788
0.897
0.655
--

Overall accuracy 0.893

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
0.8811
0.0283
0.0552
0.0327
0.0028
0.0000
building
0.1534
0.8217
0.0178
0.0067
0.0003
0.0000
low_veg
0.0421
0.0260
0.7513
0.1805
0.0002
0.0000
tree
0.0130
0.0060
0.1140
0.8670
0.0000
0.0000
car
0.3587
0.0043
0.0161
0.0055
0.6128
0.0026
clutter
--
--
--
--
--
--
Precision/Correctness
0.722
0.919
0.803
0.846
0.855
--
Recall/Completeness
0.881
0.822
0.751
0.867
0.613
--
F1
0.794
0.868
0.776
0.856
0.714
--

Overall accuracy 0.825

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9091
0.0255
0.0471
0.0173
0.0010
0.0000
building
0.1614
0.8289
0.0076
0.0018
0.0003
0.0000
low_veg
0.0397
0.0175
0.7842
0.1585
0.0001
0.0000
tree
0.0077
0.0020
0.0971
0.8931
0.0000
0.0000
car
0.2744
0.0053
0.0153
0.0011
0.7004
0.0035
clutter
--
--
--
--
--
--
Precision/Correctness
0.736
0.945
0.834
0.874
0.918
--
Recall/Completeness
0.909
0.829
0.784
0.893
0.700
--
F1
0.814
0.883
0.808
0.884
0.794
--

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_area14_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8411
0.0436
0.0822
0.0305
0.0027
0.0000
building
0.0126
0.9568
0.0260
0.0046
0.0001
0.0000
low_veg
0.0734
0.0563
0.7284
0.1407
0.0010
0.0002
tree
0.0313
0.0119
0.1248
0.8320
0.0000
0.0000
car
0.3590
0.0085
0.0043
0.0099
0.6179
0.0004
clutter
--
--
--
--
--
--
Precision/Correctness
0.857
0.856
0.740
0.865
0.650
--
Recall/Completeness
0.841
0.957
0.728
0.832
0.618
--
F1
0.849
0.904
0.734
0.848
0.633
--

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
0.8595
0.0391
0.0785
0.0209
0.0020
0.0000
building
0.0075
0.9760
0.0152
0.0013
0.0001
0.0000
low_veg
0.0669
0.0418
0.7615
0.1286
0.0010
0.0003
tree
0.0224
0.0051
0.1127
0.8598
0.0000
0.0000
car
0.2822
0.0091
0.0042
0.0057
0.6983
0.0004
clutter
--
--
--
--
--
--
Precision/Correctness
0.885
0.889
0.762
0.887
0.654
--
Recall/Completeness
0.860
0.976
0.761
0.860
0.698
--
F1
0.872
0.930
0.762
0.873
0.675
--

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8237
0.0358
0.0892
0.0490
0.0023
0.0001
building
0.0264
0.9331
0.0293
0.0111
0.0001
0.0000
low_veg
0.0244
0.0159
0.8372
0.1218
0.0000
0.0007
tree
0.0256
0.0051
0.1254
0.8438
0.0000
0.0000
car
0.3658
0.0072
0.0285
0.0063
0.5708
0.0213
clutter
--
--
--
--
--
--
Precision/Correctness
0.852
0.932
0.792
0.871
0.763
--
Recall/Completeness
0.824
0.933
0.837
0.844
0.571
--
F1
0.837
0.932
0.814
0.857
0.653
--

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8581
0.0319
0.0789
0.0302
0.0009
0.0000
building
0.0228
0.9557
0.0168
0.0046
0.0001
0.0000
low_veg
0.0212
0.0104
0.8605
0.1072
0.0000
0.0007
tree
0.0185
0.0019
0.1108
0.8688
0.0000
0.0000
car
0.3172
0.0062
0.0242
0.0030
0.6254
0.0240
clutter
--
--
--
--
--
--
Precision/Correctness
0.880
0.952
0.821
0.896
0.858
--
Recall/Completeness
0.858
0.956
0.861
0.869
0.625
--
F1
0.869
0.954
0.840
0.882
0.724
--

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
0.8582
0.0300
0.0572
0.0504
0.0043
0.0000
building
0.0287
0.9273
0.0315
0.0124
0.0001
0.0000
low_veg
0.0631
0.0166
0.7531
0.1668
0.0004
0.0000
tree
0.0093
0.0015
0.0623
0.9269
0.0000
0.0000
car
0.3733
0.0186
0.0442
0.0308
0.5314
0.0018
clutter
0.8149
0.1200
0.0569
0.0024
0.0058
0.0000
Precision/Correctness
0.796
0.920
0.768
0.895
0.659
0.000
Recall/Completeness
0.858
0.927
0.753
0.927
0.531
0.000
F1
0.826
0.924
0.761
0.911
0.588
-nan

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8873
0.0231
0.0545
0.0320
0.0032
0.0000
building
0.0252
0.9423
0.0257
0.0067
0.0001
0.0000
low_veg
0.0548
0.0070
0.7999
0.1378
0.0005
0.0000
tree
0.0041
0.0003
0.0502
0.9453
0.0000
0.0000
car
0.3225
0.0187
0.0392
0.0236
0.5938
0.0024
clutter
0.8204
0.1193
0.0538
0.0004
0.0060
0.0000
Precision/Correctness
0.816
0.943
0.796
0.923
0.655
0.000
Recall/Completeness
0.887
0.942
0.800
0.945
0.594
0.000
F1
0.850
0.943
0.798
0.934
0.623
-nan

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: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9029
0.0219
0.0381
0.0322
0.0047
0.0003
building
0.0360
0.9412
0.0173
0.0052
0.0003
0.0000
low_veg
0.1016
0.0337
0.6993
0.1653
0.0002
0.0000
tree
0.0275
0.0049
0.0672
0.9003
0.0001
0.0000
car
0.5058
0.0118
0.0118
0.0128
0.4576
0.0002
clutter
--
--
--
--
--
--
Precision/Correctness
0.877
0.916
0.802
0.853
0.754
--
Recall/Completeness
0.903
0.941
0.699
0.900
0.458
--
F1
0.890
0.929
0.747
0.876
0.569
--

Overall accuracy 0.863

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
0.9329
0.0149
0.0317
0.0170
0.0032
0.0003
building
0.0288
0.9595
0.0095
0.0019
0.0002
0.0000
low_veg
0.0977
0.0210
0.7406
0.1404
0.0002
0.0000
tree
0.0130
0.0018
0.0510
0.9341
0.0001
0.0000
car
0.4434
0.0131
0.0105
0.0046
0.5281
0.0002
clutter
--
--
--
--
--
--
Precision/Correctness
0.905
0.948
0.844
0.889
0.792
--
Recall/Completeness
0.933
0.960
0.741
0.934
0.528
--
F1
0.919
0.954
0.789
0.911
0.634
--

Overall accuracy 0.897

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area24_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8282
0.0368
0.0934
0.0387
0.0030
0.0000
building
0.0679
0.8791
0.0358
0.0169
0.0003
0.0000
low_veg
0.0272
0.0211
0.8797
0.0719
0.0001
0.0000
tree
0.0249
0.0086
0.1436
0.8229
0.0000
0.0000
car
0.4251
0.0039
0.0088
0.0070
0.5545
0.0005
clutter
--
--
--
--
--
--
Precision/Correctness
0.791
0.911
0.864
0.845
0.796
--
Recall/Completeness
0.828
0.879
0.880
0.823
0.555
--
F1
0.809
0.895
0.872
0.834
0.654
--

Overall accuracy 0.856

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
0.8581
0.0278
0.0848
0.0276
0.0018
0.0000
building
0.0645
0.8927
0.0292
0.0133
0.0003
0.0000
low_veg
0.0238
0.0165
0.8999
0.0598
0.0001
0.0000
tree
0.0185
0.0048
0.1231
0.8536
0.0000
0.0000
car
0.3496
0.0045
0.0052
0.0034
0.6367
0.0007
clutter
--
--
--
--
--
--
Precision/Correctness
0.815
0.934
0.886
0.871
0.829
--
Recall/Completeness
0.858
0.893
0.900
0.854
0.637
--
F1
0.836
0.913
0.893
0.862
0.720
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9413
0.0249
0.0195
0.0072
0.0071
0.0000
building
0.0632
0.9125
0.0146
0.0086
0.0009
0.0001
low_veg
0.1009
0.0255
0.6787
0.1943
0.0005
0.0000
tree
0.0164
0.0067
0.0550
0.9217
0.0001
0.0000
car
0.4790
0.0182
0.0058
0.0056
0.4905
0.0010
clutter
0.7689
0.0420
0.1139
0.0096
0.0646
0.0010
Precision/Correctness
0.831
0.957
0.885
0.680
0.848
0.040
Recall/Completeness
0.941
0.912
0.679
0.922
0.490
0.001
F1
0.883
0.934
0.768
0.782
0.621
0.002

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
0.9661
0.0152
0.0133
0.0027
0.0027
0.0000
building
0.0517
0.9322
0.0088
0.0064
0.0008
0.0001
low_veg
0.0817
0.0120
0.7224
0.1835
0.0004
0.0000
tree
0.0088
0.0029
0.0378
0.9506
0.0000
0.0000
car
0.4088
0.0212
0.0036
0.0047
0.5606
0.0012
clutter
0.7904
0.0286
0.1046
0.0049
0.0701
0.0014
Precision/Correctness
0.868
0.977
0.923
0.709
0.912
0.046
Recall/Completeness
0.966
0.932
0.722
0.951
0.561
0.001
F1
0.914
0.954
0.810
0.812
0.694
0.003

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_area29_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9343
0.0219
0.0262
0.0130
0.0045
0.0001
building
0.0219
0.9608
0.0098
0.0075
0.0000
0.0000
low_veg
0.0808
0.0268
0.5993
0.2928
0.0003
0.0000
tree
0.0052
0.0041
0.0320
0.9587
0.0000
0.0000
car
0.4461
0.0104
0.0087
0.0100
0.5225
0.0023
clutter
0.6402
0.1972
0.0299
0.0117
0.1024
0.0184
Precision/Correctness
0.901
0.930
0.829
0.672
0.583
0.901
Recall/Completeness
0.934
0.961
0.599
0.959
0.523
0.018
F1
0.917
0.945
0.696
0.790
0.551
0.036

Overall accuracy 0.862

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9562
0.0151
0.0198
0.0058
0.0032
0.0000
building
0.0160
0.9742
0.0055
0.0043
0.0000
0.0000
low_veg
0.0600
0.0121
0.6453
0.2823
0.0002
0.0000
tree
0.0019
0.0020
0.0227
0.9734
0.0000
0.0000
car
0.3801
0.0099
0.0071
0.0067
0.5934
0.0028
clutter
0.6408
0.1998
0.0215
0.0047
0.1129
0.0203
Precision/Correctness
0.924
0.952
0.871
0.709
0.577
0.925
Recall/Completeness
0.956
0.974
0.645
0.973
0.593
0.020
F1
0.940
0.963
0.741
0.820
0.585
0.040

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_area31_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9638
0.0128
0.0140
0.0041
0.0051
0.0002
building
0.2593
0.7214
0.0130
0.0029
0.0033
0.0002
low_veg
0.0438
0.0011
0.9142
0.0407
0.0003
0.0000
tree
0.0431
0.0030
0.0946
0.8583
0.0010
0.0000
car
0.3685
0.1777
0.0125
0.0059
0.4330
0.0024
clutter
0.5653
0.1714
0.1268
0.0073
0.0698
0.0595
Precision/Correctness
0.764
0.919
0.848
0.916
0.743
0.905
Recall/Completeness
0.964
0.721
0.914
0.858
0.433
0.059
F1
0.853
0.808
0.880
0.886
0.547
0.112

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9761
0.0097
0.0105
0.0009
0.0027
0.0001
building
0.2555
0.7352
0.0059
0.0006
0.0027
0.0002
low_veg
0.0253
0.0002
0.9410
0.0334
0.0001
0.0000
tree
0.0172
0.0006
0.0757
0.9062
0.0004
0.0000
car
0.3020
0.2030
0.0074
0.0023
0.4826
0.0028
clutter
0.5269
0.1880
0.1389
0.0024
0.0751
0.0687
Precision/Correctness
0.785
0.933
0.884
0.946
0.769
0.927
Recall/Completeness
0.976
0.735
0.941
0.906
0.483
0.069
F1
0.870
0.822
0.912
0.926
0.593
0.128

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_area33_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9753
0.0099
0.0072
0.0031
0.0037
0.0008
building
0.1796
0.7577
0.0284
0.0315
0.0025
0.0002
low_veg
0.0893
0.0137
0.8514
0.0443
0.0014
0.0000
tree
0.0411
0.0052
0.1483
0.8052
0.0001
0.0000
car
0.4795
0.0194
0.0080
0.0041
0.4874
0.0016
clutter
0.6079
0.0650
0.0034
0.0002
0.1129
0.2106
Precision/Correctness
0.775
0.955
0.830
0.831
0.424
0.967
Recall/Completeness
0.975
0.758
0.851
0.805
0.487
0.211
F1
0.864
0.845
0.841
0.818
0.454
0.346

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_area33_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9859
0.0076
0.0032
0.0008
0.0024
0.0001
building
0.1760
0.7654
0.0254
0.0307
0.0022
0.0002
low_veg
0.0571
0.0067
0.8941
0.0408
0.0013
0.0000
tree
0.0177
0.0008
0.1369
0.8444
0.0001
0.0000
car
0.4055
0.0247
0.0053
0.0023
0.5603
0.0019
clutter
0.5792
0.0653
0.0014
0.0000
0.1196
0.2344
Precision/Correctness
0.797
0.968
0.855
0.846
0.416
0.990
Recall/Completeness
0.986
0.765
0.894
0.844
0.560
0.234
F1
0.882
0.855
0.874
0.845
0.478
0.379

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8878
0.0357
0.0571
0.0181
0.0014
0.0000
building
0.0564
0.9082
0.0284
0.0059
0.0011
0.0000
low_veg
0.0828
0.0138
0.6812
0.2221
0.0002
0.0000
tree
0.0432
0.0057
0.1132
0.8377
0.0002
0.0001
car
0.4449
0.0258
0.0014
0.0042
0.5198
0.0039
clutter
--
--
--
--
--
--
Precision/Correctness
0.820
0.946
0.777
0.762
0.848
--
Recall/Completeness
0.888
0.908
0.681
0.838
0.520
--
F1
0.852
0.927
0.726
0.798
0.644
--

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_area35_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9240
0.0299
0.0400
0.0056
0.0005
0.0000
building
0.0447
0.9344
0.0186
0.0012
0.0010
0.0000
low_veg
0.0642
0.0062
0.7141
0.2154
0.0002
0.0000
tree
0.0184
0.0015
0.0902
0.8898
0.0001
0.0000
car
0.3573
0.0282
0.0002
0.0006
0.6083
0.0055
clutter
--
--
--
--
--
--
Precision/Correctness
0.868
0.966
0.830
0.789
0.882
--
Recall/Completeness
0.924
0.934
0.714
0.890
0.608
--
F1
0.895
0.950
0.768
0.836
0.720
--

Overall accuracy 0.866

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9076
0.0273
0.0464
0.0161
0.0026
0.0000
building
0.0275
0.9263
0.0391
0.0065
0.0006
0.0000
low_veg
0.0370
0.0152
0.8579
0.0898
0.0001
0.0000
tree
0.0261
0.0072
0.1137
0.8529
0.0000
0.0000
car
0.3989
0.0431
0.0345
0.0059
0.5162
0.0014
clutter
--
--
--
--
--
--
Precision/Correctness
0.903
0.946
0.843
0.842
0.850
--
Recall/Completeness
0.908
0.926
0.858
0.853
0.516
--
F1
0.905
0.936
0.850
0.848
0.642
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9347
0.0243
0.0333
0.0063
0.0014
0.0000
building
0.0198
0.9512
0.0254
0.0031
0.0006
0.0000
low_veg
0.0214
0.0086
0.8965
0.0734
0.0000
0.0000
tree
0.0110
0.0044
0.0915
0.8931
0.0000
0.0000
car
0.3144
0.0537
0.0213
0.0012
0.6083
0.0012
clutter
--
--
--
--
--
--
Precision/Correctness
0.942
0.961
0.882
0.881
0.894
--
Recall/Completeness
0.935
0.951
0.897
0.893
0.608
--
F1
0.938
0.956
0.889
0.887
0.724
--

Overall accuracy 0.918

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image