Detailled semantic labeling (2D) result


Name Y. Liu et al.
Affiliation Chinese Academy of Sciences, China
Abbreviation CASIA2
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
0.911
0.030
0.038
0.016
0.004
0.000
building
0.031
0.948
0.016
0.004
0.001
0.000
low_veg
0.052
0.025
0.797
0.126
0.000
0.000
tree
0.022
0.007
0.095
0.876
0.000
0.000
car
0.241
0.014
0.008
0.006
0.730
0.000
clutter
0.366
0.272
0.016
0.007
0.032
0.307
Precision/Correctness
0.892
0.932
0.825
0.858
0.845
0.968
Recall/Completeness
0.911
0.948
0.797
0.876
0.730
0.307
F1
0.902
0.940
0.811
0.867
0.783
0.466

Overall accuracy 0.881

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
0.938
0.021
0.031
0.007
0.002
0.000
building
0.024
0.964
0.010
0.002
0.000
0.000
low_veg
0.039
0.014
0.836
0.111
0.000
0.000
tree
0.010
0.002
0.078
0.910
0.000
0.000
car
0.131
0.013
0.005
0.003
0.848
0.000
clutter
0.329
0.284
0.012
0.002
0.033
0.339
Precision/Correctness
0.925
0.956
0.859
0.888
0.886
0.982
Recall/Completeness
0.938
0.964
0.836
0.910
0.848
0.339
F1
0.932
0.960
0.847
0.899
0.867
0.504

Overall accuracy 0.911
=============================================================

Tile top_mosaic_09cm_area2_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9472
0.0174
0.0210
0.0100
0.0044
0.0000
building
0.0274
0.9588
0.0117
0.0021
0.0000
0.0000
low_veg
0.0620
0.0299
0.7593
0.1487
0.0001
0.0000
tree
0.0304
0.0089
0.0961
0.8633
0.0002
0.0011
car
0.2556
0.0064
0.0000
0.0014
0.7361
0.0005
clutter
0.0134
0.0000
0.0000
0.0170
0.0000
0.9696
Precision/Correctness
0.877
0.940
0.854
0.843
0.906
0.985
Recall/Completeness
0.947
0.959
0.759
0.863
0.736
0.970
F1
0.911
0.949
0.804
0.853
0.812
0.977

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_area2_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9714
0.0122
0.0112
0.0025
0.0028
0.0000
building
0.0214
0.9716
0.0065
0.0005
0.0000
0.0000
low_veg
0.0448
0.0163
0.7961
0.1428
0.0000
0.0000
tree
0.0140
0.0034
0.0829
0.8993
0.0001
0.0004
car
0.1124
0.0035
0.0000
0.0006
0.8828
0.0007
clutter
0.0138
0.0000
0.0000
0.0035
0.0000
0.9827
Precision/Correctness
0.920
0.967
0.884
0.865
0.937
0.994
Recall/Completeness
0.971
0.972
0.796
0.899
0.883
0.983
F1
0.945
0.969
0.838
0.882
0.909
0.988

Overall accuracy 0.911

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
0.9443
0.0374
0.0097
0.0043
0.0042
0.0000
building
0.0548
0.9366
0.0068
0.0012
0.0005
0.0000
low_veg
0.1252
0.0853
0.5302
0.2586
0.0007
0.0000
tree
0.0489
0.0168
0.0793
0.8549
0.0001
0.0000
car
0.2784
0.0130
0.0006
0.0026
0.7054
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.866
0.971
0.693
0.752
0.868
--
Recall/Completeness
0.944
0.937
0.530
0.855
0.705
--
F1
0.903
0.954
0.601
0.800
0.778
--

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_area4_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9733
0.0164
0.0068
0.0015
0.0020
0.0000
building
0.0428
0.9507
0.0056
0.0003
0.0005
0.0000
low_veg
0.0853
0.0532
0.5765
0.2850
0.0000
0.0000
tree
0.0199
0.0066
0.0744
0.8991
0.0000
0.0000
car
0.1605
0.0074
0.0000
0.0011
0.8310
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.899
0.987
0.725
0.779
0.913
--
Recall/Completeness
0.973
0.951
0.576
0.899
0.831
--
F1
0.934
0.969
0.642
0.835
0.870
--

Overall accuracy 0.938

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
0.8971
0.0646
0.0234
0.0097
0.0051
0.0000
building
0.0261
0.9625
0.0088
0.0024
0.0002
0.0000
low_veg
0.0844
0.0443
0.7412
0.1296
0.0005
0.0000
tree
0.0161
0.0062
0.0635
0.9140
0.0002
0.0000
car
0.2754
0.0224
0.0002
0.0016
0.7004
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.895
0.932
0.744
0.939
0.887
--
Recall/Completeness
0.897
0.963
0.741
0.914
0.700
--
F1
0.896
0.947
0.743
0.926
0.783
--

Overall accuracy 0.904

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
0.9380
0.0422
0.0157
0.0024
0.0018
0.0000
building
0.0204
0.9729
0.0058
0.0008
0.0001
0.0000
low_veg
0.0676
0.0221
0.8001
0.1101
0.0001
0.0000
tree
0.0058
0.0016
0.0509
0.9417
0.0000
0.0000
car
0.1658
0.0145
0.0000
0.0001
0.8196
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.931
0.962
0.787
0.960
0.953
--
Recall/Completeness
0.938
0.973
0.800
0.942
0.820
--
F1
0.934
0.967
0.794
0.951
0.881
--

Overall accuracy 0.936

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
0.8978
0.0236
0.0671
0.0080
0.0034
0.0000
building
0.0284
0.9516
0.0161
0.0039
0.0000
0.0000
low_veg
0.0495
0.0140
0.8832
0.0525
0.0008
0.0000
tree
0.0269
0.0069
0.0939
0.8721
0.0002
0.0000
car
0.2577
0.0090
0.0122
0.0017
0.7194
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.912
0.948
0.785
0.938
0.906
--
Recall/Completeness
0.898
0.952
0.883
0.872
0.719
--
F1
0.905
0.950
0.832
0.904
0.802
--

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_area8_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9242
0.0149
0.0575
0.0020
0.0015
0.0000
building
0.0196
0.9707
0.0083
0.0014
0.0000
0.0000
low_veg
0.0330
0.0051
0.9174
0.0443
0.0002
0.0000
tree
0.0103
0.0029
0.0714
0.9154
0.0000
0.0000
car
0.1410
0.0081
0.0066
0.0003
0.8439
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.949
0.973
0.827
0.958
0.955
--
Recall/Completeness
0.924
0.971
0.917
0.915
0.844
--
F1
0.937
0.972
0.870
0.936
0.896
--

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_area10_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8748
0.0499
0.0522
0.0176
0.0053
0.0002
building
0.0342
0.9409
0.0145
0.0079
0.0025
0.0000
low_veg
0.0822
0.0309
0.7544
0.1325
0.0000
0.0000
tree
0.0161
0.0084
0.0893
0.8860
0.0001
0.0001
car
0.2559
0.0177
0.0003
0.0010
0.7251
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.922
0.895
0.762
0.851
0.663
--
Recall/Completeness
0.875
0.941
0.754
0.886
0.725
--
F1
0.898
0.918
0.758
0.868
0.693
--

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
0.9069
0.0369
0.0453
0.0068
0.0039
0.0002
building
0.0282
0.9539
0.0109
0.0048
0.0022
0.0000
low_veg
0.0783
0.0161
0.8008
0.1048
0.0000
0.0000
tree
0.0061
0.0038
0.0697
0.9204
0.0000
0.0000
car
0.1737
0.0188
0.0001
0.0000
0.8074
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.940
0.928
0.799
0.895
0.677
--
Recall/Completeness
0.907
0.954
0.801
0.920
0.807
--
F1
0.923
0.941
0.800
0.908
0.736
--

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_area12_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8803
0.0283
0.0587
0.0307
0.0020
0.0000
building
0.0215
0.9580
0.0150
0.0056
0.0000
0.0000
low_veg
0.0360
0.0187
0.7764
0.1688
0.0000
0.0000
tree
0.0087
0.0068
0.0991
0.8854
0.0000
0.0000
car
0.2268
0.0000
0.0000
0.0145
0.7586
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.869
0.939
0.825
0.857
0.929
--
Recall/Completeness
0.880
0.958
0.776
0.885
0.759
--
F1
0.874
0.948
0.800
0.871
0.835
--

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_area12_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9076
0.0243
0.0520
0.0156
0.0004
0.0000
building
0.0198
0.9722
0.0064
0.0017
0.0000
0.0000
low_veg
0.0325
0.0086
0.8129
0.1460
0.0000
0.0000
tree
0.0035
0.0019
0.0819
0.9126
0.0000
0.0000
car
0.1008
0.0000
0.0000
0.0040
0.8952
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.898
0.966
0.855
0.885
0.984
--
Recall/Completeness
0.908
0.972
0.813
0.913
0.895
--
F1
0.903
0.969
0.833
0.899
0.937
--

Overall accuracy 0.895

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
0.8510
0.0304
0.0817
0.0298
0.0069
0.0002
building
0.0090
0.9716
0.0151
0.0042
0.0000
0.0001
low_veg
0.0783
0.0529
0.7317
0.1363
0.0007
0.0000
tree
0.0263
0.0122
0.1217
0.8397
0.0000
0.0000
car
0.1390
0.0000
0.0000
0.0055
0.8555
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.864
0.874
0.750
0.869
0.563
--
Recall/Completeness
0.851
0.972
0.732
0.840
0.856
--
F1
0.858
0.920
0.741
0.854
0.679
--

Overall accuracy 0.838

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
0.8713
0.0252
0.0781
0.0191
0.0061
0.0002
building
0.0048
0.9869
0.0071
0.0011
0.0000
0.0001
low_veg
0.0731
0.0368
0.7642
0.1251
0.0007
0.0000
tree
0.0182
0.0053
0.1087
0.8679
0.0000
0.0000
car
0.0427
0.0000
0.0000
0.0008
0.9565
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.889
0.911
0.772
0.891
0.541
--
Recall/Completeness
0.871
0.987
0.764
0.868
0.957
--
F1
0.880
0.947
0.768
0.879
0.691
--

Overall accuracy 0.864

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image


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

Tile top_mosaic_09cm_area16_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8349
0.0213
0.0909
0.0506
0.0023
0.0000
building
0.0201
0.9530
0.0189
0.0080
0.0000
0.0000
low_veg
0.0124
0.0125
0.8586
0.1154
0.0000
0.0011
tree
0.0242
0.0047
0.1020
0.8690
0.0000
0.0000
car
0.1168
0.0024
0.0307
0.0028
0.8474
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.887
0.951
0.823
0.880
0.832
--
Recall/Completeness
0.835
0.953
0.859
0.869
0.847
--
F1
0.860
0.952
0.840
0.874
0.840
--

Overall accuracy 0.876

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
0.8699
0.0162
0.0830
0.0304
0.0006
0.0000
building
0.0158
0.9751
0.0066
0.0024
0.0000
0.0000
low_veg
0.0084
0.0059
0.8846
0.0999
0.0000
0.0012
tree
0.0168
0.0011
0.0877
0.8944
0.0000
0.0000
car
0.0380
0.0000
0.0260
0.0001
0.9358
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.919
0.974
0.851
0.905
0.940
--
Recall/Completeness
0.870
0.975
0.885
0.894
0.936
--
F1
0.894
0.975
0.867
0.899
0.938
--

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_area20_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8769
0.0287
0.0474
0.0443
0.0027
0.0000
building
0.0044
0.9846
0.0057
0.0053
0.0000
0.0000
low_veg
0.0592
0.0354
0.7386
0.1669
0.0000
0.0000
tree
0.0089
0.0022
0.0473
0.9417
0.0000
0.0000
car
0.1257
0.0013
0.0403
0.0152
0.8176
0.0000
clutter
0.0000
0.9996
0.0002
0.0001
0.0000
0.0000
Precision/Correctness
0.915
0.799
0.826
0.902
0.860
-nan
Recall/Completeness
0.877
0.985
0.739
0.942
0.818
0.000
F1
0.896
0.882
0.780
0.921
0.838
-nan

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_area20_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9112
0.0169
0.0454
0.0257
0.0008
0.0000
building
0.0023
0.9944
0.0023
0.0010
0.0000
0.0000
low_veg
0.0523
0.0224
0.7883
0.1370
0.0000
0.0000
tree
0.0037
0.0003
0.0361
0.9599
0.0000
0.0000
car
0.0464
0.0001
0.0358
0.0070
0.9107
0.0000
clutter
0.0000
1.0000
0.0000
0.0000
0.0000
0.0000
Precision/Correctness
0.940
0.822
0.855
0.929
0.943
-nan
Recall/Completeness
0.911
0.994
0.788
0.960
0.911
0.000
F1
0.925
0.900
0.820
0.944
0.927
-nan

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_area22_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9008
0.0285
0.0387
0.0266
0.0054
0.0000
building
0.0196
0.9680
0.0086
0.0038
0.0001
0.0000
low_veg
0.0862
0.0432
0.7462
0.1242
0.0002
0.0000
tree
0.0231
0.0062
0.0734
0.8972
0.0001
0.0000
car
0.2250
0.0000
0.0075
0.0168
0.7507
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.905
0.898
0.811
0.883
0.821
--
Recall/Completeness
0.901
0.968
0.746
0.897
0.751
--
F1
0.903
0.932
0.777
0.890
0.784
--

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
0.9325
0.0194
0.0323
0.0129
0.0030
0.0000
building
0.0141
0.9800
0.0046
0.0013
0.0000
0.0000
low_veg
0.0819
0.0245
0.7967
0.0968
0.0000
0.0000
tree
0.0092
0.0019
0.0559
0.9329
0.0000
0.0000
car
0.1223
0.0000
0.0038
0.0051
0.8688
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.932
0.939
0.850
0.920
0.877
--
Recall/Completeness
0.932
0.980
0.797
0.933
0.869
--
F1
0.932
0.959
0.823
0.926
0.873
--

Overall accuracy 0.914

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
0.8343
0.0314
0.1023
0.0273
0.0047
0.0000
building
0.0559
0.8904
0.0436
0.0099
0.0000
0.0002
low_veg
0.0197
0.0161
0.8978
0.0663
0.0000
0.0000
tree
0.0172
0.0070
0.1634
0.8122
0.0001
0.0000
car
0.1959
0.0000
0.0034
0.0049
0.7958
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.837
0.928
0.851
0.862
0.805
--
Recall/Completeness
0.834
0.890
0.898
0.812
0.796
--
F1
0.835
0.909
0.874
0.836
0.800
--

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_area24_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8663
0.0213
0.0920
0.0176
0.0028
0.0000
building
0.0527
0.9037
0.0369
0.0067
0.0000
0.0000
low_veg
0.0163
0.0115
0.9158
0.0563
0.0000
0.0000
tree
0.0106
0.0036
0.1389
0.8468
0.0001
0.0000
car
0.1084
0.0000
0.0001
0.0017
0.8898
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.862
0.952
0.875
0.885
0.851
--
Recall/Completeness
0.866
0.904
0.916
0.847
0.890
--
F1
0.864
0.927
0.895
0.866
0.870
--

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_area27_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9275
0.0412
0.0180
0.0066
0.0067
0.0000
building
0.0366
0.9517
0.0085
0.0030
0.0002
0.0000
low_veg
0.0790
0.0473
0.6916
0.1819
0.0003
0.0000
tree
0.0103
0.0084
0.0567
0.9246
0.0000
0.0000
car
0.2843
0.0083
0.0034
0.0055
0.6986
0.0000
clutter
0.3850
0.4673
0.1030
0.0036
0.0411
0.0000
Precision/Correctness
0.882
0.931
0.902
0.704
0.905
-nan
Recall/Completeness
0.928
0.952
0.692
0.925
0.699
0.000
F1
0.904
0.941
0.783
0.799
0.789
-nan

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: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9577
0.0264
0.0115
0.0024
0.0020
0.0000
building
0.0287
0.9645
0.0053
0.0013
0.0002
0.0000
low_veg
0.0612
0.0255
0.7430
0.1701
0.0001
0.0000
tree
0.0034
0.0023
0.0412
0.9531
0.0000
0.0000
car
0.1617
0.0048
0.0015
0.0041
0.8279
0.0000
clutter
0.3493
0.5097
0.0848
0.0008
0.0554
0.0000
Precision/Correctness
0.918
0.960
0.935
0.735
0.960
-nan
Recall/Completeness
0.958
0.964
0.743
0.953
0.828
0.000
F1
0.938
0.962
0.828
0.830
0.889
-nan

Overall accuracy 0.914

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
0.9394
0.0206
0.0219
0.0129
0.0052
0.0000
building
0.0258
0.9597
0.0092
0.0053
0.0000
0.0000
low_veg
0.0605
0.0228
0.6703
0.2449
0.0015
0.0000
tree
0.0047
0.0025
0.0336
0.9593
0.0000
0.0000
car
0.2224
0.0036
0.0006
0.0131
0.7586
0.0017
clutter
0.4835
0.4034
0.0422
0.0065
0.0313
0.0332
Precision/Correctness
0.922
0.914
0.856
0.707
0.751
0.973
Recall/Completeness
0.939
0.960
0.670
0.959
0.759
0.033
F1
0.930
0.937
0.752
0.814
0.755
0.064

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_area29_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9619
0.0136
0.0152
0.0057
0.0037
0.0000
building
0.0198
0.9728
0.0050
0.0024
0.0000
0.0000
low_veg
0.0376
0.0096
0.7194
0.2320
0.0015
0.0000
tree
0.0009
0.0003
0.0242
0.9746
0.0000
0.0000
car
0.1264
0.0006
0.0000
0.0100
0.8613
0.0018
clutter
0.4687
0.4241
0.0327
0.0029
0.0345
0.0372
Precision/Correctness
0.944
0.936
0.896
0.747
0.764
0.982
Recall/Completeness
0.962
0.973
0.719
0.975
0.861
0.037
F1
0.953
0.954
0.798
0.846
0.810
0.072

Overall accuracy 0.907

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
0.9720
0.0090
0.0081
0.0043
0.0063
0.0003
building
0.0218
0.9676
0.0062
0.0022
0.0016
0.0005
low_veg
0.0301
0.0039
0.9310
0.0338
0.0006
0.0005
tree
0.0424
0.0042
0.0928
0.8596
0.0010
0.0000
car
0.1517
0.0481
0.0067
0.0066
0.7869
0.0000
clutter
0.6082
0.0848
0.0277
0.0117
0.0540
0.2135
Precision/Correctness
0.902
0.967
0.893
0.925
0.852
0.955
Recall/Completeness
0.972
0.968
0.931
0.860
0.787
0.214
F1
0.935
0.967
0.912
0.891
0.818
0.349

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_area31_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9856
0.0063
0.0048
0.0009
0.0023
0.0001
building
0.0114
0.9851
0.0016
0.0002
0.0013
0.0004
low_veg
0.0129
0.0015
0.9585
0.0271
0.0000
0.0000
tree
0.0179
0.0008
0.0730
0.9081
0.0002
0.0000
car
0.0624
0.0550
0.0020
0.0027
0.8779
0.0000
clutter
0.5763
0.0895
0.0252
0.0051
0.0576
0.2463
Precision/Correctness
0.936
0.976
0.928
0.955
0.893
0.981
Recall/Completeness
0.986
0.985
0.958
0.908
0.878
0.246
F1
0.960
0.980
0.943
0.931
0.885
0.394

Overall accuracy 0.947

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
0.9745
0.0155
0.0040
0.0034
0.0019
0.0007
building
0.0389
0.9311
0.0275
0.0022
0.0003
0.0000
low_veg
0.0710
0.0237
0.8748
0.0298
0.0008
0.0000
tree
0.0395
0.0101
0.1329
0.8173
0.0001
0.0000
car
0.2199
0.0071
0.0069
0.0016
0.7646
0.0000
clutter
0.4950
0.0804
0.0005
0.0017
0.0488
0.3735
Precision/Correctness
0.870
0.947
0.851
0.928
0.733
0.986
Recall/Completeness
0.974
0.931
0.875
0.817
0.765
0.374
F1
0.919
0.939
0.863
0.869
0.748
0.542

Overall accuracy 0.895

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
0.9877
0.0099
0.0009
0.0008
0.0006
0.0001
building
0.0327
0.9414
0.0252
0.0005
0.0002
0.0000
low_veg
0.0395
0.0131
0.9209
0.0259
0.0005
0.0000
tree
0.0147
0.0039
0.1230
0.8585
0.0000
0.0000
car
0.1137
0.0019
0.0037
0.0004
0.8803
0.0000
clutter
0.4515
0.0750
0.0000
0.0011
0.0537
0.4187
Precision/Correctness
0.903
0.966
0.872
0.952
0.740
0.997
Recall/Completeness
0.988
0.941
0.921
0.858
0.880
0.419
F1
0.943
0.954
0.896
0.903
0.804
0.590

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_area35_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.8646
0.0615
0.0557
0.0165
0.0017
0.0000
building
0.0275
0.9508
0.0163
0.0054
0.0000
0.0000
low_veg
0.0502
0.0146
0.7378
0.1973
0.0000
0.0000
tree
0.0380
0.0081
0.1112
0.8426
0.0001
0.0000
car
0.2784
0.0040
0.0000
0.0084
0.7092
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.876
0.923
0.805
0.783
0.921
--
Recall/Completeness
0.865
0.951
0.738
0.843
0.709
--
F1
0.870
0.937
0.770
0.812
0.801
--

Overall accuracy 0.850

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
0.9047
0.0558
0.0352
0.0040
0.0003
0.0000
building
0.0189
0.9729
0.0077
0.0006
0.0000
0.0000
low_veg
0.0352
0.0042
0.7710
0.1895
0.0000
0.0000
tree
0.0135
0.0024
0.0858
0.8983
0.0000
0.0000
car
0.1689
0.0037
0.0000
0.0024
0.8250
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.925
0.947
0.861
0.812
0.982
--
Recall/Completeness
0.905
0.973
0.771
0.898
0.825
--
F1
0.915
0.960
0.814
0.853
0.897
--

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_area38_class.tif, reference set: full_reference


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9070
0.0219
0.0503
0.0178
0.0027
0.0002
building
0.0268
0.9254
0.0401
0.0061
0.0017
0.0000
low_veg
0.0385
0.0133
0.8588
0.0889
0.0005
0.0000
tree
0.0234
0.0035
0.1175
0.8555
0.0001
0.0000
car
0.3124
0.0024
0.0454
0.0093
0.6306
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.906
0.958
0.837
0.842
0.821
--
Recall/Completeness
0.907
0.925
0.859
0.856
0.631
--
F1
0.907
0.941
0.847
0.849
0.713
--

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_area38_class.tif, reference set: no_boundary


↓ predicted || reference →
imp_surf
building
low_veg
tree
car
clutter
imp_surf
0.9354
0.0184
0.0373
0.0074
0.0013
0.0002
building
0.0195
0.9520
0.0247
0.0023
0.0015
0.0000
low_veg
0.0220
0.0085
0.8976
0.0717
0.0003
0.0000
tree
0.0082
0.0007
0.0944
0.8967
0.0000
0.0000
car
0.2014
0.0001
0.0366
0.0025
0.7594
0.0000
clutter
--
--
--
--
--
--
Precision/Correctness
0.946
0.971
0.877
0.883
0.871
--
Recall/Completeness
0.935
0.952
0.898
0.897
0.759
--
F1
0.941
0.962
0.887
0.890
0.811
--

Overall accuracy 0.921

Red/green image, indicating wrongly classified pixels


Original, resized true ortho image for better comparison


Classified image