Detailled semantic labeling (2D) result

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 0.926 0.015 0.032 0.016 0.002 0.009 building 0.017 0.966 0.008 0.003 0.000 0.005 low_veg 0.039 0.009 0.882 0.059 0.000 0.011 tree 0.033 0.005 0.125 0.833 0.002 0.003 car 0.068 0.003 0.001 0.013 0.907 0.009 clutter 0.269 0.113 0.148 0.020 0.010 0.439 Precision/Correctness 0.902 0.948 0.822 0.881 0.915 0.747 Recall/Completeness 0.926 0.966 0.882 0.833 0.907 0.439 F1 0.914 0.957 0.851 0.856 0.911 0.553 Overall accuracy 0.888	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 0.944 0.011 0.026 0.011 0.000 0.007 building 0.014 0.974 0.006 0.002 0.000 0.005 low_veg 0.029 0.005 0.906 0.051 0.000 0.010 tree 0.023 0.004 0.109 0.861 0.002 0.002 car 0.008 0.003 0.000 0.008 0.972 0.009 clutter 0.266 0.099 0.133 0.017 0.010 0.476 Precision/Correctness 0.923 0.963 0.847 0.906 0.949 0.774 Recall/Completeness 0.944 0.974 0.906 0.861 0.972 0.476 F1 0.934 0.968 0.876 0.883 0.961 0.590 Overall accuracy 0.910
============================================================= Tile top_potsdam_2_13_class.tif, reference set: full_reference ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9003 0.0247 0.0491 0.0093 0.0021 0.0146 building 0.0128 0.9545 0.0222 0.0033 0.0004 0.0068 low_veg 0.0282 0.0099 0.9106 0.0343 0.0002 0.0167 tree 0.0382 0.0102 0.1339 0.8126 0.0016 0.0035 car 0.0753 0.0106 0.0024 0.0066 0.8992 0.0059 clutter 0.1191 0.1932 0.1415 0.0159 0.0116 0.5187 Precision/Correctness 0.906 0.920 0.847 0.932 0.895 0.481 Recall/Completeness 0.900 0.955 0.911 0.813 0.899 0.519 F1 0.903 0.937 0.877 0.868 0.897 0.499 Overall accuracy 0.886 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image	============================================================= Tile top_potsdam_2_13_class.tif, reference set: no_boundary ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9220 0.0196 0.0403 0.0054 0.0002 0.0124 building 0.0095 0.9633 0.0186 0.0023 0.0003 0.0059 low_veg 0.0209 0.0062 0.9321 0.0252 0.0001 0.0155 tree 0.0293 0.0083 0.1118 0.8467 0.0011 0.0027 car 0.0169 0.0131 0.0007 0.0021 0.9621 0.0052 clutter 0.0833 0.1907 0.1017 0.0100 0.0149 0.5995 Precision/Correctness 0.933 0.942 0.873 0.954 0.934 0.472 Recall/Completeness 0.922 0.963 0.932 0.847 0.962 0.600 F1 0.927 0.952 0.902 0.897 0.948 0.528 Overall accuracy 0.911 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image
============================================================= Tile top_potsdam_2_14_class.tif, reference set: full_reference ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9028 0.0084 0.0731 0.0093 0.0012 0.0051 building 0.0053 0.9809 0.0092 0.0029 0.0000 0.0017 low_veg 0.0130 0.0039 0.9564 0.0249 0.0000 0.0017 tree 0.0189 0.0036 0.2022 0.7739 0.0008 0.0006 car 0.0572 0.0037 0.0013 0.0090 0.9283 0.0005 clutter 0.0952 0.2966 0.1675 0.0381 0.0376 0.3649 Precision/Correctness 0.913 0.928 0.824 0.958 0.901 0.418 Recall/Completeness 0.903 0.981 0.956 0.774 0.928 0.365 F1 0.908 0.954 0.885 0.856 0.914 0.390 Overall accuracy 0.883 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image	============================================================= Tile top_potsdam_2_14_class.tif, reference set: no_boundary ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9312 0.0053 0.0536 0.0057 0.0000 0.0041 building 0.0033 0.9895 0.0046 0.0016 0.0000 0.0011 low_veg 0.0085 0.0020 0.9689 0.0194 0.0000 0.0013 tree 0.0146 0.0028 0.1841 0.7975 0.0005 0.0005 car 0.0019 0.0024 0.0000 0.0031 0.9924 0.0002 clutter 0.0470 0.3362 0.1199 0.0366 0.0448 0.4155 Precision/Correctness 0.938 0.952 0.843 0.969 0.937 0.432 Recall/Completeness 0.931 0.990 0.969 0.798 0.992 0.416 F1 0.935 0.970 0.901 0.875 0.964 0.423 Overall accuracy 0.902 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image
============================================================= Tile top_potsdam_3_13_class.tif, reference set: full_reference ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9043 0.0210 0.0431 0.0162 0.0033 0.0121 building 0.0065 0.9810 0.0067 0.0026 0.0000 0.0032 low_veg 0.0505 0.0187 0.8746 0.0483 0.0001 0.0078 tree 0.0274 0.0066 0.1020 0.8592 0.0026 0.0022 car 0.0637 0.0015 0.0005 0.0125 0.9100 0.0119 clutter 0.2763 0.1634 0.2393 0.0249 0.0064 0.2897 Precision/Correctness 0.893 0.942 0.829 0.923 0.907 0.520 Recall/Completeness 0.904 0.981 0.875 0.859 0.910 0.290 F1 0.899 0.961 0.851 0.890 0.908 0.372 Overall accuracy 0.891 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image	============================================================= Tile top_potsdam_3_13_class.tif, reference set: no_boundary ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9310 0.0139 0.0339 0.0100 0.0002 0.0110 building 0.0042 0.9879 0.0039 0.0016 0.0000 0.0024 low_veg 0.0392 0.0094 0.9081 0.0375 0.0000 0.0057 tree 0.0192 0.0050 0.0843 0.8879 0.0019 0.0017 car 0.0066 0.0010 0.0000 0.0075 0.9716 0.0133 clutter 0.2882 0.1589 0.2085 0.0197 0.0076 0.3171 Precision/Correctness 0.921 0.963 0.860 0.947 0.954 0.526 Recall/Completeness 0.931 0.988 0.908 0.888 0.972 0.317 F1 0.926 0.976 0.883 0.916 0.963 0.396 Overall accuracy 0.920 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image
============================================================= Tile top_potsdam_3_14_class.tif, reference set: full_reference ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.8924 0.0204 0.0551 0.0193 0.0035 0.0094 building 0.0114 0.9803 0.0036 0.0026 0.0000 0.0020 low_veg 0.0239 0.0093 0.9254 0.0369 0.0000 0.0045 tree 0.0194 0.0043 0.1958 0.7758 0.0012 0.0035 car 0.0584 0.0006 0.0006 0.0255 0.9077 0.0072 clutter 0.2042 0.1812 0.1906 0.0182 0.0050 0.4009 Precision/Correctness 0.900 0.941 0.785 0.928 0.901 0.723 Recall/Completeness 0.892 0.980 0.925 0.776 0.908 0.401 F1 0.896 0.961 0.849 0.845 0.904 0.516 Overall accuracy 0.876 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image	============================================================= Tile top_potsdam_3_14_class.tif, reference set: no_boundary ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9221 0.0150 0.0419 0.0130 0.0004 0.0077 building 0.0091 0.9869 0.0009 0.0017 0.0000 0.0014 low_veg 0.0143 0.0041 0.9508 0.0278 0.0000 0.0030 tree 0.0142 0.0033 0.1784 0.8002 0.0009 0.0029 car 0.0072 0.0003 0.0000 0.0155 0.9704 0.0066 clutter 0.1938 0.1746 0.1572 0.0124 0.0060 0.4561 Precision/Correctness 0.928 0.960 0.809 0.951 0.951 0.766 Recall/Completeness 0.922 0.987 0.951 0.800 0.970 0.456 F1 0.925 0.973 0.874 0.869 0.961 0.572 Overall accuracy 0.901 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image
============================================================= Tile top_potsdam_4_13_class.tif, reference set: full_reference ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9391 0.0093 0.0251 0.0227 0.0015 0.0024 building 0.0146 0.9699 0.0074 0.0042 0.0002 0.0037 low_veg 0.0458 0.0059 0.8417 0.1013 0.0001 0.0051 tree 0.0259 0.0020 0.0452 0.9248 0.0016 0.0005 car 0.0715 0.0002 0.0004 0.0236 0.9024 0.0019 clutter 0.5207 0.1108 0.1601 0.0626 0.0141 0.1316 Precision/Correctness 0.846 0.966 0.821 0.810 0.921 0.761 Recall/Completeness 0.939 0.970 0.842 0.925 0.902 0.132 F1 0.890 0.968 0.831 0.864 0.912 0.224 Overall accuracy 0.878 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image	============================================================= Tile top_potsdam_4_13_class.tif, reference set: no_boundary ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9572 0.0062 0.0189 0.0160 0.0002 0.0015 building 0.0096 0.9813 0.0031 0.0028 0.0001 0.0031 low_veg 0.0336 0.0015 0.8698 0.0909 0.0000 0.0041 tree 0.0151 0.0010 0.0320 0.9507 0.0010 0.0003 car 0.0050 0.0000 0.0000 0.0146 0.9788 0.0016 clutter 0.5503 0.0994 0.1421 0.0543 0.0153 0.1387 Precision/Correctness 0.870 0.976 0.860 0.846 0.938 0.796 Recall/Completeness 0.957 0.981 0.870 0.951 0.979 0.139 F1 0.911 0.979 0.865 0.895 0.958 0.236 Overall accuracy 0.903 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image
============================================================= Tile top_potsdam_4_14_class.tif, reference set: full_reference ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.8953 0.0170 0.0568 0.0130 0.0030 0.0148 building 0.0161 0.9675 0.0040 0.0053 0.0001 0.0070 low_veg 0.0791 0.0135 0.8037 0.0765 0.0002 0.0270 tree 0.0296 0.0041 0.1222 0.8355 0.0020 0.0064 car 0.0627 0.0042 0.0004 0.0214 0.8960 0.0153 clutter 0.4131 0.0394 0.1909 0.0129 0.0005 0.3432 Precision/Correctness 0.767 0.941 0.764 0.869 0.941 0.724 Recall/Completeness 0.895 0.967 0.804 0.836 0.896 0.343 F1 0.826 0.954 0.783 0.852 0.918 0.466 Overall accuracy 0.823 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image	============================================================= Tile top_potsdam_4_14_class.tif, reference set: no_boundary ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9127 0.0126 0.0530 0.0082 0.0002 0.0134 building 0.0129 0.9753 0.0018 0.0042 0.0000 0.0058 low_veg 0.0667 0.0084 0.8281 0.0699 0.0000 0.0269 tree 0.0210 0.0029 0.1034 0.8654 0.0016 0.0057 car 0.0175 0.0038 0.0000 0.0147 0.9489 0.0152 clutter 0.4236 0.0353 0.1796 0.0103 0.0003 0.3509 Precision/Correctness 0.781 0.957 0.783 0.892 0.979 0.746 Recall/Completeness 0.913 0.975 0.828 0.865 0.949 0.351 F1 0.842 0.966 0.805 0.878 0.964 0.477 Overall accuracy 0.842 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image
============================================================= Tile top_potsdam_4_15_class.tif, reference set: full_reference ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9286 0.0279 0.0311 0.0059 0.0023 0.0043 building 0.0204 0.9646 0.0069 0.0021 0.0003 0.0058 low_veg 0.0619 0.0176 0.8792 0.0339 0.0003 0.0071 tree 0.0642 0.0057 0.1205 0.7998 0.0067 0.0032 car 0.0823 0.0080 0.0010 0.0049 0.9014 0.0024 clutter 0.3303 0.1978 0.1596 0.0110 0.0090 0.2924 Precision/Correctness 0.898 0.940 0.814 0.922 0.913 0.635 Recall/Completeness 0.929 0.965 0.879 0.800 0.901 0.292 F1 0.913 0.952 0.846 0.857 0.907 0.400 Overall accuracy 0.896 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image	============================================================= Tile top_potsdam_4_15_class.tif, reference set: no_boundary ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9440 0.0229 0.0263 0.0036 0.0003 0.0029 building 0.0185 0.9708 0.0043 0.0014 0.0002 0.0048 low_veg 0.0401 0.0097 0.9163 0.0286 0.0000 0.0053 tree 0.0467 0.0041 0.0985 0.8434 0.0049 0.0024 car 0.0131 0.0095 0.0000 0.0015 0.9738 0.0020 clutter 0.3363 0.1914 0.1502 0.0088 0.0091 0.3043 Precision/Correctness 0.924 0.956 0.844 0.944 0.948 0.663 Recall/Completeness 0.944 0.971 0.916 0.843 0.974 0.304 F1 0.934 0.963 0.879 0.891 0.961 0.417 Overall accuracy 0.921 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image
============================================================= Tile top_potsdam_5_13_class.tif, reference set: full_reference ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9317 0.0105 0.0285 0.0216 0.0016 0.0062 building 0.0141 0.9749 0.0043 0.0033 0.0000 0.0034 low_veg 0.0700 0.0105 0.8382 0.0756 0.0000 0.0058 tree 0.0450 0.0031 0.0986 0.8492 0.0025 0.0015 car 0.0865 0.0020 0.0002 0.0076 0.8984 0.0052 clutter 0.2045 0.2227 0.1781 0.0135 0.0617 0.3195 Precision/Correctness 0.907 0.955 0.848 0.819 0.891 0.697 Recall/Completeness 0.932 0.975 0.838 0.849 0.898 0.319 F1 0.919 0.965 0.843 0.834 0.895 0.438 Overall accuracy 0.895 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image	============================================================= Tile top_potsdam_5_13_class.tif, reference set: no_boundary ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9500 0.0058 0.0234 0.0158 0.0001 0.0048 building 0.0102 0.9820 0.0027 0.0025 0.0000 0.0027 low_veg 0.0646 0.0061 0.8591 0.0662 0.0000 0.0040 tree 0.0297 0.0017 0.0848 0.8806 0.0020 0.0011 car 0.0091 0.0023 0.0000 0.0050 0.9784 0.0052 clutter 0.1652 0.2098 0.1763 0.0111 0.0774 0.3603 Precision/Correctness 0.931 0.971 0.876 0.849 0.904 0.725 Recall/Completeness 0.950 0.982 0.859 0.881 0.978 0.360 F1 0.941 0.977 0.867 0.864 0.940 0.481 Overall accuracy 0.919 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image
============================================================= Tile top_potsdam_5_14_class.tif, reference set: full_reference ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9377 0.0052 0.0243 0.0240 0.0012 0.0075 building 0.0084 0.9847 0.0035 0.0019 0.0000 0.0015 low_veg 0.0512 0.0086 0.8405 0.0860 0.0001 0.0136 tree 0.0412 0.0047 0.0750 0.8717 0.0034 0.0040 car 0.0845 0.0009 0.0010 0.0136 0.8737 0.0264 clutter 0.2075 0.1966 0.1493 0.0379 0.0034 0.4054 Precision/Correctness 0.872 0.930 0.846 0.846 0.935 0.824 Recall/Completeness 0.938 0.985 0.840 0.872 0.874 0.405 F1 0.904 0.957 0.843 0.859 0.903 0.543 Overall accuracy 0.877 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image	============================================================= Tile top_potsdam_5_14_class.tif, reference set: no_boundary ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9555 0.0027 0.0174 0.0178 0.0001 0.0065 building 0.0065 0.9896 0.0014 0.0013 0.0000 0.0011 low_veg 0.0368 0.0057 0.8651 0.0801 0.0000 0.0124 tree 0.0309 0.0036 0.0653 0.8943 0.0025 0.0035 car 0.0064 0.0006 0.0001 0.0075 0.9565 0.0290 clutter 0.1915 0.1750 0.1352 0.0359 0.0033 0.4590 Precision/Correctness 0.902 0.948 0.873 0.869 0.956 0.847 Recall/Completeness 0.955 0.990 0.865 0.894 0.956 0.459 F1 0.928 0.968 0.869 0.882 0.956 0.595 Overall accuracy 0.901 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image
============================================================= Tile top_potsdam_5_15_class.tif, reference set: full_reference ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9119 0.0252 0.0395 0.0188 0.0012 0.0035 building 0.0158 0.9669 0.0074 0.0052 0.0002 0.0045 low_veg 0.0406 0.0040 0.8160 0.1197 0.0001 0.0197 tree 0.0298 0.0033 0.0647 0.8987 0.0022 0.0014 car 0.0946 0.0028 0.0016 0.0128 0.8778 0.0104 clutter 0.2692 0.1352 0.1227 0.0122 0.0030 0.4577 Precision/Correctness 0.913 0.921 0.776 0.794 0.958 0.847 Recall/Completeness 0.912 0.967 0.816 0.899 0.878 0.458 F1 0.913 0.943 0.795 0.843 0.916 0.594 Overall accuracy 0.878 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image	============================================================= Tile top_potsdam_5_15_class.tif, reference set: no_boundary ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9262 0.0220 0.0345 0.0140 0.0005 0.0028 building 0.0117 0.9747 0.0055 0.0039 0.0002 0.0040 low_veg 0.0259 0.0008 0.8414 0.1133 0.0000 0.0186 tree 0.0200 0.0018 0.0542 0.9214 0.0016 0.0011 car 0.0102 0.0031 0.0002 0.0096 0.9659 0.0110 clutter 0.2634 0.1054 0.1099 0.0101 0.0030 0.5082 Precision/Correctness 0.934 0.940 0.801 0.823 0.971 0.867 Recall/Completeness 0.926 0.975 0.841 0.921 0.966 0.508 F1 0.930 0.957 0.821 0.869 0.968 0.641 Overall accuracy 0.900 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image
============================================================= Tile top_potsdam_6_13_class.tif, reference set: full_reference ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9419 0.0100 0.0207 0.0149 0.0016 0.0110 building 0.0358 0.9389 0.0186 0.0011 0.0003 0.0051 low_veg 0.0151 0.0047 0.9446 0.0342 0.0000 0.0014 tree 0.0545 0.0022 0.0683 0.8725 0.0012 0.0013 car 0.0611 0.0000 0.0004 0.0039 0.9279 0.0066 clutter 0.3307 0.0182 0.3187 0.0274 0.0167 0.2884 Precision/Correctness 0.931 0.976 0.892 0.877 0.908 0.499 Recall/Completeness 0.942 0.939 0.945 0.873 0.928 0.288 F1 0.936 0.957 0.918 0.875 0.918 0.365 Overall accuracy 0.920 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image	============================================================= Tile top_potsdam_6_13_class.tif, reference set: no_boundary ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9584 0.0079 0.0152 0.0091 0.0001 0.0092 building 0.0318 0.9494 0.0140 0.0007 0.0002 0.0040 low_veg 0.0087 0.0027 0.9600 0.0277 0.0000 0.0009 tree 0.0371 0.0012 0.0566 0.9031 0.0009 0.0011 car 0.0101 0.0000 0.0001 0.0018 0.9815 0.0064 clutter 0.3324 0.0149 0.3334 0.0256 0.0181 0.2755 Precision/Correctness 0.947 0.983 0.912 0.908 0.948 0.509 Recall/Completeness 0.958 0.949 0.960 0.903 0.982 0.276 F1 0.952 0.966 0.935 0.905 0.965 0.358 Overall accuracy 0.938 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image
============================================================= Tile top_potsdam_6_14_class.tif, reference set: full_reference ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9522 0.0119 0.0189 0.0108 0.0039 0.0024 building 0.0323 0.9592 0.0054 0.0009 0.0004 0.0018 low_veg 0.0429 0.0092 0.8728 0.0681 0.0004 0.0066 tree 0.0346 0.0076 0.0622 0.8899 0.0048 0.0010 car 0.0469 0.0030 0.0009 0.0083 0.9362 0.0046 clutter 0.2960 0.3001 0.1364 0.0148 0.0104 0.2423 Precision/Correctness 0.930 0.954 0.881 0.844 0.925 0.672 Recall/Completeness 0.952 0.959 0.873 0.890 0.936 0.242 F1 0.941 0.957 0.877 0.866 0.931 0.356 Overall accuracy 0.919 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image	============================================================= Tile top_potsdam_6_14_class.tif, reference set: no_boundary ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9681 0.0084 0.0139 0.0076 0.0002 0.0019 building 0.0283 0.9670 0.0027 0.0005 0.0002 0.0013 low_veg 0.0273 0.0049 0.9014 0.0614 0.0000 0.0050 tree 0.0236 0.0063 0.0495 0.9162 0.0036 0.0009 car 0.0023 0.0022 0.0004 0.0045 0.9863 0.0043 clutter 0.2848 0.3330 0.1151 0.0115 0.0092 0.2464 Precision/Correctness 0.947 0.963 0.909 0.872 0.974 0.712 Recall/Completeness 0.968 0.967 0.901 0.916 0.986 0.246 F1 0.958 0.965 0.905 0.894 0.980 0.366 Overall accuracy 0.938 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image
============================================================= Tile top_potsdam_6_15_class.tif, reference set: full_reference ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9303 0.0087 0.0317 0.0098 0.0048 0.0147 building 0.0105 0.9619 0.0193 0.0034 0.0002 0.0047 low_veg 0.0306 0.0051 0.8737 0.0509 0.0009 0.0388 tree 0.0351 0.0040 0.1396 0.8139 0.0042 0.0033 car 0.0482 0.0013 0.0027 0.0121 0.9269 0.0088 clutter 0.1010 0.0288 0.1190 0.0062 0.0040 0.7410 Precision/Correctness 0.946 0.972 0.741 0.908 0.908 0.771 Recall/Completeness 0.930 0.962 0.874 0.814 0.927 0.741 F1 0.938 0.967 0.802 0.859 0.917 0.756 Overall accuracy 0.900 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image	============================================================= Tile top_potsdam_6_15_class.tif, reference set: no_boundary ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9502 0.0060 0.0246 0.0063 0.0003 0.0127 building 0.0060 0.9708 0.0164 0.0025 0.0001 0.0041 low_veg 0.0149 0.0035 0.9015 0.0422 0.0001 0.0378 tree 0.0261 0.0026 0.1221 0.8433 0.0032 0.0027 car 0.0038 0.0011 0.0003 0.0083 0.9781 0.0083 clutter 0.0932 0.0282 0.1060 0.0047 0.0037 0.7642 Precision/Correctness 0.965 0.980 0.766 0.933 0.966 0.795 Recall/Completeness 0.950 0.971 0.901 0.843 0.978 0.764 F1 0.957 0.975 0.828 0.886 0.972 0.779 Overall accuracy 0.921 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image
============================================================= Tile top_potsdam_7_13_class.tif, reference set: full_reference ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9398 0.0122 0.0088 0.0246 0.0017 0.0130 building 0.0265 0.9419 0.0050 0.0031 0.0002 0.0233 low_veg 0.0430 0.0075 0.7753 0.1655 0.0001 0.0087 tree 0.0250 0.0031 0.0915 0.8782 0.0011 0.0011 car 0.0479 0.0007 0.0002 0.0221 0.9199 0.0092 clutter 0.1272 0.0131 0.0428 0.0013 0.0116 0.8040 Precision/Correctness 0.950 0.963 0.826 0.695 0.860 0.851 Recall/Completeness 0.940 0.942 0.775 0.878 0.920 0.804 F1 0.945 0.952 0.800 0.776 0.889 0.827 Overall accuracy 0.906 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image	============================================================= Tile top_potsdam_7_13_class.tif, reference set: no_boundary ↓ predicted || reference → imp_surf building low_veg tree car clutter imp_surf 0.9546 0.0091 0.0061 0.0177 0.0004 0.0122 building 0.0230 0.9476 0.0040 0.0022 0.0002 0.0230 low_veg 0.0312 0.0032 0.7992 0.1585 0.0000 0.0079 tree 0.0169 0.0016 0.0844 0.8954 0.0007 0.0010 car 0.0053 0.0004 0.0000 0.0147 0.9714 0.0082 clutter 0.1182 0.0117 0.0404 0.0009 0.0109 0.8180 Precision/Correctness 0.959 0.973 0.847 0.725 0.891 0.860 Recall/Completeness 0.955 0.948 0.799 0.895 0.971 0.818 F1 0.957 0.960 0.823 0.801 0.929 0.838 Overall accuracy 0.921 Red/green image, indicating wrongly classified pixels Original, resized true ortho image for better comparison Classified image