<html>
<head>
<style><!--
.hmmessage P
{
margin:0px;
padding:0px
}
body.hmmessage
{
font-size: 10pt;
font-family:Verdana
}
--></style>
</head>
<body class='hmmessage'>
<BR> Hi everyone,<BR>
<BR>
I have the results using MaximumStepLength=1.0. They are the following:<BR>
<BR>
<BR>0 13424.5 [-0.255739, 1000, 1000, -359.703, 937.071]<BR>1 13440 [-0.105437, 1000.24, 1000.06, -359.83, 938.021]<BR>2 13608.1 [0.260693, 1000.31, 1000.12, -360.377, 938.768]<BR>3 12618.7 [1.00548, 1000.35, 999.961, -360.981, 938.539]<BR>4 5568.94 [0.838302, 1001.01, 999.763, -360.833, 937.844]<BR>5 6426.39 [0.774628, 1001.63, 999.818, -360.458, 937.167]<BR>6 6522.07 [0.560376, 1002.22, 999.828, -360.151, 936.451]<BR>7 9178.12 [1.3541, 1002.45, 1000.01, -359.721, 936.138]<BR>8 8177.64 [0.894182, 1002.37, 999.881, -359.842, 936.126]<BR>9 5849.88 [0.903255, 1002.53, 999.89, -359.752, 935.96]<BR>10 5747.78 [0.915017, 1002.7, 999.896, -359.663, 935.795]<BR>11 5619.39 [0.929918, 1002.86, 999.902, -359.575, 935.63]<BR>12 5478.51 [0.938114, 1003.03, 999.901, -359.493, 935.463]<BR>13 5423.5 [0.936299, 1003.2, 999.893, -359.417, 935.294]<BR>14 5418.88 [0.933334, 1003.36, 999.885, -359.341, 935.124]<BR>15 5418.56 [0.931768, 1003.53, 999.878, -359.264, 934.955]<BR>16 5412.59 [0.932801, 1003.7, 999.874, -359.185, 934.787]<BR>17 5395.26 [0.9304, 1003.86, 999.867, -359.108, 934.618]<BR>18 5394.36 [0.933178, 1004.03, 999.867, -359.025, 934.451]<BR>19 5368.52 [0.928946, 1004.2, 999.858, -358.95, 934.281]<BR>20 5377.68 [0.933322, 1004.37, 999.86, -358.865, 934.115]<BR>21 5343.16 [0.929035, 1004.53, 999.851, -358.79, 933.945]<BR>22 5351.25 [0.930128, 1004.7, 999.852, -358.706, 933.779]<BR>23 5332.42 [0.927707, 1004.87, 999.848, -358.627, 933.611]<BR>24 5334.16 [0.932723, 1005.03, 999.848, -358.543, 933.444]<BR>25 5297.11 [0.926336, 1005.2, 999.839, -358.469, 933.274]<BR>26 5318.14 [0.931818, 1005.37, 999.843, -358.381, 933.109]<BR>27 5276.94 [0.924499, 1005.53, 999.835, -358.306, 932.939]<BR>28 5305.93 [0.929325, 1005.7, 999.84, -358.217, 932.776]<BR>29 5263.45 [0.923843, 1005.87, 999.835, -358.139, 932.606]<BR>30 5285.18 [0.928561, 1006.03, 999.841, -358.05, 932.443]<BR>31 5243.59 [0.92428, 1006.2, 999.836, -357.971, 932.274]<BR>32 5257.08 [0.927006, 1006.36, 999.841, -357.883, 932.109]<BR>33 5227.63 [0.923191, 1006.53, 999.84, -357.802, 931.942]<BR>34 5240.49 [0.926061, 1006.7, 999.845, -357.713, 931.777]<BR>35 5209.68 [0.923094, 1006.86, 999.845, -357.63, 931.611]<BR>36 5216.99 [0.924202, 1007.03, 999.849, -357.542, 931.446]<BR>37 5197.77 [0.924582, 1007.2, 999.855, -357.454, 931.282]<BR>38 5183.73 [0.922479, 1007.36, 999.857, -357.368, 931.116]<BR>39 5185.34 [0.921585, 1007.53, 999.861, -357.282, 930.95]<BR>40 5178.81 [0.920225, 1007.7, 999.864, -357.195, 930.785]<BR>41 5175.71 [0.91913, 1007.86, 999.866, -357.11, 930.618]<BR>42 5170.77 [0.918411, 1008.03, 999.869, -357.024, 930.452]<BR>43 5163.45 [0.917265, 1008.19, 999.871, -356.939, 930.285]<BR>44 5158.96 [0.916038, 1008.36, 999.872, -356.856, 930.118]<BR>45 5154.82 [0.915832, 1008.52, 999.873, -356.772, 929.951]<BR>46 5144.11 [0.914556, 1008.69, 999.874, -356.688, 929.783]<BR>47 5140.7 [0.913587, 1008.86, 999.874, -356.604, 929.615]<BR>48 5135.16 [0.911516, 1009.02, 999.875, -356.521, 929.447]<BR>49 5135.62 [0.909232, 1009.19, 999.875, -356.439, 929.279]<BR>50 5138.31 [0.908764, 1009.35, 999.876, -356.356, 929.11]<BR>51 5128.93 [0.907338, 1009.52, 999.876, -356.273, 928.942]<BR>52 5126.22 [0.906331, 1009.68, 999.877, -356.19, 928.773]<BR>53 5120.44 [0.904524, 1009.85, 999.878, -356.106, 928.605]<BR>54 5119.71 [0.902134, 1010.01, 999.88, -356.021, 928.437]<BR>55 5122.26 [0.899496, 1010.17, 999.883, -355.937, 928.268]<BR>56 5127.11 [0.894635, 1010.34, 999.886, -355.852, 928.1]<BR>57 5150.51 [0.893125, 1010.5, 999.889, -355.767, 927.931]<BR>58 5148.8 [0.891511, 1010.67, 999.891, -355.682, 927.762]<BR>59 5148.34 [0.890091, 1010.83, 999.894, -355.598, 927.592]<BR>60 5146.79 [0.888496, 1010.99, 999.896, -355.514, 927.422]<BR>61 5146.05 [0.884676, 1011.15, 999.898, -355.431, 927.252]<BR>62 5157.42 [0.876514, 1011.32, 999.899, -355.348, 927.081]<BR>63 5196.03 [0.86541, 1011.48, 999.904, -355.262, 926.911]<BR>64 5233.59 [0.849185, 1011.64, 999.913, -355.173, 926.742]<BR>65 5307.1 [0.832145, 1011.8, 999.919, -355.086, 926.57]<BR>66 5446.17 [0.79583, 1011.95, 999.924, -355.003, 926.397]<BR>67 5484.05 [0.746317, 1012.1, 999.929, -354.923, 926.221]<BR>68 5891.16 [0.763592, 1012.25, 999.952, -354.82, 926.047]<BR>69 5628.46 [0.757405, 1012.4, 999.972, -354.722, 925.873]<BR>70 5698.92 [0.752866, 1012.54, 999.993, -354.621, 925.699]<BR>71 5747.29 [0.760902, 1012.69, 1000.02, -354.519, 925.525]<BR>72 5625.99 [0.762871, 1012.84, 1000.04, -354.416, 925.353]<BR>73 5584.26 [0.763336, 1012.98, 1000.06, -354.316, 925.179]<BR>74 5564.81 [0.763232, 1013.13, 1000.08, -354.217, 925.005]<BR>75 5553.12 [0.764406, 1013.28, 1000.1, -354.116, 924.831]<BR>76 5524.3 [0.761778, 1013.43, 1000.12, -354.018, 924.656]<BR>77 5548.29 [0.763268, 1013.58, 1000.14, -353.916, 924.484]<BR>78 5514.28 [0.764806, 1013.72, 1000.17, -353.816, 924.31]<BR>79 5479.82 [0.763734, 1013.87, 1000.18, -353.718, 924.136]<BR>80 5482.17 [0.765634, 1014.02, 1000.21, -353.616, 923.963]<BR>81 5442.63 [0.763086, 1014.17, 1000.23, -353.517, 923.789]<BR>82 5465.47 [0.763811, 1014.31, 1000.25, -353.416, 923.616]<BR>83 5442.63 [0.764442, 1014.46, 1000.27, -353.315, 923.442]<BR>84 5421.21 [0.766169, 1014.61, 1000.29, -353.213, 923.27]<BR>85 5384.8 [0.765418, 1014.76, 1000.32, -353.111, 923.097]<BR>86 5382.46 [0.766298, 1014.91, 1000.34, -353.009, 922.925]<BR>87 5357.5 [0.765325, 1015.05, 1000.36, -352.907, 922.753]<BR>88 5358.53 [0.76659, 1015.2, 1000.39, -352.803, 922.582]<BR>89 5328.22 [0.767173, 1015.35, 1000.41, -352.701, 922.41]<BR>90 5307.05 [0.766803, 1015.5, 1000.43, -352.601, 922.236]<BR>91 5299.23 [0.768287, 1015.65, 1000.45, -352.497, 922.065]<BR>92 5265.99 [0.767005, 1015.79, 1000.48, -352.394, 921.894]<BR>93 5270.3 [0.769547, 1015.94, 1000.5, -352.29, 921.723]<BR>94 5222.54 [0.767441, 1016.09, 1000.52, -352.189, 921.55]<BR>95 5238.16 [0.770055, 1016.24, 1000.55, -352.086, 921.379]<BR>96 5188.58 [0.768803, 1016.39, 1000.57, -351.985, 921.206]<BR>97 5192.2 [0.76994, 1016.54, 1000.59, -351.882, 921.035]<BR>98 5163.03 [0.769595, 1016.68, 1000.61, -351.78, 920.863]<BR>99 5154.21 [0.77008, 1016.83, 1000.64, -351.678, 920.691]<BR>100 5134.02 [0.770968, 1016.98, 1000.66, -351.577, 920.518]<BR>101 5108.28 [0.771262, 1017.13, 1000.68, -351.474, 920.347]<BR>102 5091.1 [0.770405, 1017.28, 1000.7, -351.374, 920.174]<BR>103 5088.91 [0.771964, 1017.43, 1000.72, -351.272, 920.002]<BR>104 5053.96 [0.771432, 1017.58, 1000.75, -351.171, 919.83]<BR>105 5047.36 [0.772927, 1017.72, 1000.77, -351.07, 919.658]<BR>106 5013.01 [0.771788, 1017.87, 1000.79, -350.97, 919.485]<BR>107 5015.38 [0.773656, 1018.02, 1000.81, -350.869, 919.312]<BR>108 4975.13 [0.773429, 1018.17, 1000.83, -350.768, 919.141]<BR>109 4964.35 [0.774629, 1018.32, 1000.85, -350.667, 918.968]<BR>110 4934.23 [0.774557, 1018.47, 1000.87, -350.567, 918.796]<BR>111 4921.03 [0.774083, 1018.62, 1000.89, -350.468, 918.623]<BR>112 4913.73 [0.775189, 1018.77, 1000.91, -350.369, 918.449]<BR>113 4883.88 [0.776354, 1018.92, 1000.93, -350.271, 918.276]<BR>114 4853.41 [0.777329, 1019.07, 1000.95, -350.173, 918.102]<BR>115 4825.51 [0.777899, 1019.22, 1000.97, -350.076, 917.928]<BR>116 4803.71 [0.777606, 1019.37, 1000.99, -349.98, 917.754]<BR>117 4794 [0.777837, 1019.52, 1001, -349.883, 917.579]<BR>118 4776.07 [0.777358, 1019.67, 1001.02, -349.787, 917.405]<BR>119 4768.57 [0.777804, 1019.82, 1001.04, -349.692, 917.23]<BR>120 4748.83 [0.778854, 1019.97, 1001.05, -349.596, 917.055]<BR>121 4720.38 [0.776662, 1020.12, 1001.07, -349.502, 916.88]<BR>122 4736.96 [0.780539, 1020.27, 1001.09, -349.406, 916.705]<BR>123 4671.59 [0.776214, 1020.42, 1001.1, -349.313, 916.529]<BR>124 4717.37 [0.783392, 1020.57, 1001.12, -349.217, 916.355]<BR>125 4608.19 [0.776325, 1020.72, 1001.13, -349.125, 916.178]<BR>126 4690.07 [0.783974, 1020.87, 1001.15, -349.029, 916.004]<BR>127 4574.86 [0.776806, 1021.02, 1001.16, -348.938, 915.827]<BR>128 4656.83 [0.784622, 1021.17, 1001.18, -348.842, 915.653]<BR>129 4540.4 [0.779003, 1021.32, 1001.19, -348.751, 915.476]<BR>130 4598.92 [0.781051, 1021.47, 1001.21, -348.657, 915.301]<BR>131 4558.32 [0.778687, 1021.62, 1001.22, -348.564, 915.125]<BR>132 4579.05 [0.783813, 1021.77, 1001.24, -348.467, 914.951]<BR>133 4498.71 [0.779233, 1021.92, 1001.25, -348.377, 914.774]<BR>134 4546.08 [0.784053, 1022.07, 1001.27, -348.281, 914.6]<BR>135 4470.01 [0.778611, 1022.23, 1001.28, -348.192, 914.422]<BR>136 4531.88 [0.784019, 1022.38, 1001.3, -348.096, 914.248]<BR>137 4444.9 [0.777739, 1022.53, 1001.31, -348.007, 914.07]<BR>138 4520.88 [0.786422, 1022.68, 1001.33, -347.911, 913.896]<BR>139 4390.04 [0.780512, 1022.83, 1001.34, -347.824, 913.718]<BR>140 4454.43 [0.78564, 1022.98, 1001.35, -347.729, 913.543]<BR>141 4370.41 [0.78143, 1023.13, 1001.36, -347.642, 913.365]<BR>142 4416.33 [0.785899, 1023.28, 1001.38, -347.548, 913.189]<BR>143 4346.2 [0.778008, 1023.43, 1001.39, -347.46, 913.011]<BR>144 4445.96 [0.792399, 1023.58, 1001.4, -347.366, 912.836]<BR>145 4247.78 [0.770852, 1023.74, 1001.4, -347.288, 912.655]<BR>146 4551.69 [0.792196, 1023.88, 1001.42, -347.188, 912.483]<BR>147 4223.14 [0.77066, 1024.04, 1001.42, -347.113, 912.301]<BR>148 4533.53 [0.794099, 1024.18, 1001.45, -347.01, 912.13]<BR>149 4179.34 [0.771084, 1024.34, 1001.45, -346.935, 911.948]<BR>150 4504.82 [0.79392, 1024.48, 1001.47, -346.833, 911.777]<BR>151 4154.77 [0.770755, 1024.64, 1001.47, -346.759, 911.595]<BR>152 4489.76 [0.797811, 1024.78, 1001.49, -346.657, 911.425]<BR>153 4095.84 [0.769473, 1024.94, 1001.49, -346.584, 911.242]<BR>154 4494.92 [0.803377, 1025.08, 1001.51, -346.48, 911.074]<BR>155 4030.62 [0.770113, 1025.23, 1001.51, -346.41, 910.892]<BR>156 4461.06 [0.80425, 1025.38, 1001.53, -346.308, 910.722]<BR>157 3997.78 [0.769258, 1025.53, 1001.52, -346.238, 910.54]<BR>158 4458.95 [0.808118, 1025.68, 1001.55, -346.137, 910.371]<BR>159 3954.47 [0.76563, 1025.83, 1001.54, -346.072, 910.188]<BR>160 4524.49 [0.811616, 1025.98, 1001.56, -345.971, 910.019]<BR>161 3921.44 [0.762618, 1026.13, 1001.55, -345.908, 909.837]<BR>162 4577.13 [0.809255, 1026.27, 1001.57, -345.806, 909.669]<BR>163 3893.42 [0.765351, 1026.43, 1001.56, -345.742, 909.486]<BR>164 4496.93 [0.813156, 1026.57, 1001.59, -345.64, 909.318]<BR>165 3859.98 [0.763161, 1026.72, 1001.57, -345.58, 909.136]<BR>166 4529.88 [0.810638, 1026.87, 1001.6, -345.48, 908.966]<BR>167 3832.25 [0.765687, 1027.02, 1001.59, -345.417, 908.784]<BR>168 4453.69 [0.812472, 1027.16, 1001.61, -345.316, 908.615]<BR>169 3800.72 [0.762545, 1027.32, 1001.59, -345.256, 908.432]<BR>170 4508.67 [0.809582, 1027.46, 1001.62, -345.157, 908.262]<BR>171 3776.19 [0.769981, 1027.61, 1001.61, -345.09, 908.08]<BR>172 4317.05 [0.816268, 1027.76, 1001.63, -344.994, 907.909]<BR>173 3743.01 [0.763963, 1027.91, 1001.61, -344.938, 907.725]<BR>174 4440.61 [0.812532, 1028.06, 1001.63, -344.84, 907.555]<BR>175 3713.69 [0.763651, 1028.21, 1001.62, -344.781, 907.372]<BR>176 4429.8 [0.811437, 1028.35, 1001.64, -344.682, 907.202]<BR>177 3686.22 [0.767353, 1028.5, 1001.62, -344.62, 907.019]<BR>178 4326.97 [0.816903, 1028.65, 1001.65, -344.52, 906.85]<BR>179 3656.25 [0.766627, 1028.8, 1001.63, -344.465, 906.667]<BR>180 4327.23 [0.817175, 1028.95, 1001.65, -344.368, 906.496]<BR>181 3627.33 [0.7678, 1029.1, 1001.63, -344.312, 906.313]<BR>182 4284.39 [0.816754, 1029.24, 1001.65, -344.215, 906.142]<BR>183 3597.85 [0.768725, 1029.39, 1001.63, -344.158, 905.959]<BR>184 4248.18 [0.818375, 1029.54, 1001.65, -344.063, 905.788]<BR>185 3567.13 [0.768158, 1029.69, 1001.64, -344.007, 905.604]<BR>186 4244.64 [0.817634, 1029.84, 1001.65, -343.914, 905.432]<BR>187 3538.41 [0.768204, 1029.99, 1001.63, -343.857, 905.248]<BR>188 4227.9 [0.818152, 1030.14, 1001.65, -343.763, 905.076]<BR>189 3508.87 [0.767408, 1030.29, 1001.63, -343.707, 904.893]<BR>190 4229.26 [0.815955, 1030.43, 1001.65, -343.615, 904.72]<BR>191 3480.5 [0.770449, 1030.58, 1001.64, -343.556, 904.537]<BR>192 4145.04 [0.821881, 1030.73, 1001.65, -343.463, 904.364]<BR>193 3453.25 [0.770055, 1030.88, 1001.63, -343.408, 904.182]<BR>194 4139.2 [0.820555, 1031.03, 1001.65, -343.315, 904.009]<BR>195 3424.7 [0.768593, 1031.18, 1001.63, -343.259, 903.826]<BR>196 4155.43 [0.816269, 1031.33, 1001.65, -343.169, 903.652]<BR>197 3396.23 [0.771267, 1031.48, 1001.63, -343.111, 903.468]<BR>198 4082.3 [0.822574, 1031.63, 1001.65, -343.019, 903.296]<BR>199 3369.11 [0.769351, 1031.78, 1001.63, -342.967, 903.112]<BR>Result =<BR> Angle (radians) 0.769351<BR> Angle (degrees) 44.0806<BR> Center X = 1031.78<BR> Center Y = 1001.63<BR> Translation X = -342.967<BR> Translation Y = 903.112<BR> Iterations = 200<BR> Metric value = 3369.11<BR> Stop Condition = 5<BR>Matrix =<BR>0.718362 -0.695669<BR>0.695669 0.718362<BR>
Offset =<BR>[644.42, 467.432]<BR>
<BR>
<BR>
<BR>
Manuel.<BR>
<BR>
<BR>
<HR id=stopSpelling>
From: manueldavid.tabas@hotmail.es<BR>To: insight-users@itk.org<BR>Date: Thu, 22 Oct 2009 17:56:45 +0200<BR>Subject: Re: [Insight-users] Problem with 2D Registration (local minima)<BR><BR>
<STYLE>
.ExternalClass .ecxhmmessage P
{padding:0px;}
.ExternalClass body.ecxhmmessage
{font-size:10pt;font-family:Verdana;}
</STYLE>
<BR>Hi Bill,<BR> <BR>The optimum value for the metric that I spect is about 2000, and I normally get a metric of 4000.<BR> <BR>I will try just now to set:<BR> <BR>optimizer->SetMaximumStepLength( 0.1 );<BR> <BR>and I´ll let you know the results in a while.<BR> <BR>Thanks!<BR> <BR>Manuel.<BR> <BR>> Date: Thu, 22 Oct 2009 11:52:23 -0400<BR>> Subject: Re: [Insight-users] Problem with 2D Registration (local minima)<BR>> From: bill.lorensen@gmail.com<BR>> To: manueldavid.tabas@hotmail.es<BR>> CC: insight-users@itk.org<BR>> <BR>> The optimizer is stopping after reaching the maximum number of<BR>> iterations. This is normally not the best way to terminate an<BR>> optimization.<BR>> <BR>> What do you expect for the optimum value of the parameters?<BR>> <BR>> Perhaps you need a larger maximum step length? Or perhaps, as Luis<BR>> suggested, the starting point is too far from the optimum for this<BR>> optimizer.<BR>> <BR>> Bill<BR>> <BR>> <BR>> On Thu, Oct 22, 2009 at 10:09 AM, Manuel David Tabas Calle<BR>> <manueldavid.tabas@hotmail.es> wrote:<BR>> ><BR>> > Hi Luis,<BR>> ><BR>> > Thank you for answering me.<BR>> ><BR>> > 1) As you said, my images are from Microscopy and the spacing is about 2.85<BR>> > micrometres/pixel. The images are in a raw format and I used the following<BR>> > .mhd file for reading it:<BR>> ><BR>> ><BR>> > NDims = 2<BR>> > DimSize = 2000 2000<BR>> > ElementType = MET_UCHAR<BR>> > ElementSpacing = 1.0 1.0<BR>> > Offset = 0.0 0.0<BR>> > ElementByteOrderMSB = False<BR>> > ElementDataFile = fixed.raw<BR>> ><BR>> > I also tried to as ElementSpacing 2.83 and 0.00283 (just in case...) but the<BR>> > results weren´t good either.<BR>> ><BR>> > 2) I also try to put a relaxation factor of 0.9 as you said and I couldn´t<BR>> > see any improvement.<BR>> ><BR>> > 3) I´m attaching three .txt files with some results obtained.<BR>> > The results1.txt file shows the output of the program as you can find it on<BR>> > ITK/Examples/Registration/.<BR>> > The results2.txt file shows the output of the program with a predifined<BR>> > center in the position 1000x1000. The results3.txt file shows the output of<BR>> > the program with a predifined center in the position 1000x1000, a predefined<BR>> > angle of 0.9 radians and a predefined traslation of -340 in the X axis and<BR>> > 800 in the Y axis (that are the values for an optimal registration).<BR>> ><BR>> ><BR>> > I hope you can help me to fix my problem with that information... an be free<BR>> > to ask anything else :P.<BR>> ><BR>> > Thank you really much.<BR>> ><BR>> > Manuel.<BR>> ><BR>> ><BR>> >> Date: Wed, 21 Oct 2009 17:29:08 -0400<BR>> >> Subject: Re: [Insight-users] Problem with 2D Registration (local minima)<BR>> >> From: luis.ibanez@kitware.com<BR>> >> To: manueldavid.tabas@hotmail.es<BR>> >> CC: insight-users@itk.org<BR>> >><BR>> >> Hi Manuel,<BR>> >><BR>> >> Thanks for the detailed description of your problem.<BR>> >><BR>> >> A couple of questions:<BR>> >><BR>> >><BR>> >> 1) What is the pixel spacing of your image ?<BR>> >><BR>> >> The typical proportion of :<BR>> >><BR>> >> > const double translationScale = 1.0 / 1000.0;<BR>> >><BR>> >> is set up for images whose physical extent in<BR>> >> the units used by the spacing is in the range<BR>> >> of hundreds.<BR>> >><BR>> >> For example, a typical CT scan will have 512x512<BR>> >> pixels with a spacing close to 1mm / pixel.<BR>> >><BR>> >> Your images are probably from Microscopy or<BR>> >> from a Satellite, so I'm wondering about the units<BR>> >> that you use (microns or kilometers) and the<BR>> >> exact numeric values of the spacing.<BR>> >><BR>> >><BR>> >> 2) Could you post to the list the output of the<BR>> >> command observer that tracks the values of<BR>> >> the metric and the transform at every iteration ?<BR>> >><BR>> >><BR>> >> Something that happens often when using the<BR>> >> RegularStepGradientDescentOptimizer is that,<BR>> >> if the registration process have a couple of<BR>> >> bounces of direction in the parametric space,<BR>> >> the step length may be reduced too fast, and<BR>> >> not give the optimizer the oportunity to walk<BR>> >> far enough to get to the solution.<BR>> >><BR>> >><BR>> >> You may want to try using the method:<BR>> >><BR>> >> optimizer->SetRelaxationFactor( 0.9 );<BR>> >><BR>> >><BR>> >> This method defines the factor by which the<BR>> >> step length is reduced at every time that the<BR>> >> parametric space step changes direction.<BR>> >> This may prevent the premature shrinkage<BR>> >> of the step length (assuming that this is the<BR>> >> real problem).<BR>> >><BR>> >><BR>> >> Anothe typical issue: It may simply be that you<BR>> >> need a better initialization for your images.<BR>> >><BR>> >><BR>> >> Note that the Transform initializer in the example:<BR>> >><BR>> >> ImageRegistration6.cxx<BR>> >><BR>> >> is using the methods MomentsOn().<BR>> >><BR>> >> This may or may not be the right method to use,<BR>> >> depending on the content of your images.<BR>> >><BR>> >> For example: If the two images have different<BR>> >> modalities, or if the main features of one image<BR>> >> may not overlap to the one in the other image.<BR>> >><BR>> >> If you could post a screenshot of both images<BR>> >> we could avoid all this speculation... :-)<BR>> >><BR>> >> Or...<BR>> >> If you can share your images in a public web<BR>> >> site, that will be even better.<BR>> >><BR>> >><BR>> >><BR>> >> Thanks<BR>> >><BR>> >><BR>> >> Luis<BR>> >><BR>> >><BR>> >> ----------------------------------------------------------------<BR>> >> On Wed, Oct 21, 2009 at 11:56 AM, Manuel David Tabas Calle<BR>> >> <manueldavid.tabas@hotmail.es> wrote:<BR>> >> > Hello ITK,<BR>> >> ><BR>> >> > I´m working with ITK to do registration of 2D images and I think I reach<BR>> >> > a<BR>> >> > local minima that makes a poor registration.<BR>> >> ><BR>> >> > I´m using 2D images 2000x2000 pixels in what the moving image needs a<BR>> >> > rotation and a translation. So I reproduced the ImageRegistration6.cxx<BR>> >> > to do<BR>> >> > that. But the result I got is not good enough so I think that the<BR>> >> > MeanSquares metric reaches a local minima. The value of the metric is<BR>> >> > about<BR>> >> > 4000 (aprox.), and I run a test using as input parameters the optimal<BR>> >> > ones<BR>> >> > (that is, the result I think i should obtain) and the metric returns a<BR>> >> > value<BR>> >> > of 2500 (aprox.).<BR>> >> ><BR>> >> > The framework I use is:<BR>> >> ><BR>> >> > itkImageRegistrationMethod<BR>> >> > itkMeanSquaresImageToImageMetric<BR>> >> > itkLinearInterpolateImageFunction<BR>> >> > itkRegularStepGradientDescentOptimizer<BR>> >> > itkCenteredRigid2DTransform<BR>> >> ><BR>> >> > And the parameters of my registration are:<BR>> >> ><BR>> >> > const double translationScale = 1.0 / 1000.0;<BR>> >> > optimizerScales[0] = 1.0;<BR>> >> > optimizerScales[1] = translationScale;<BR>> >> > optimizerScales[2] = translationScale;<BR>> >> > optimizerScales[3] = translationScale;<BR>> >> > optimizerScales[4] = translationScale;<BR>> >> ><BR>> >> > optimizer->SetMaximumStepLength( 0.1 );<BR>> >> > optimizer->SetMinimumStepLength( 0.001 );<BR>> >> > optimizer->SetNumberOfIterations( 200 );<BR>> >> ><BR>> >> ><BR>> >> > The scales are well defined and I tried to vary the parameters of the<BR>> >> > optimizer but the results were similar, so I don´t know what to do next.<BR>> >> > Any<BR>> >> > idea??<BR>> >> ><BR>> >> > Thank you in advance, I hope you can have more ideas to try.<BR>> >> ><BR>> >> ><BR>> >> ><BR>> >> > Manuel.<BR>> >> ><BR>> >> ><BR>> >> ><BR>> >> ><BR>> >> > _____________________________________<BR>> >> > Powered by www.kitware.com<BR>> >> ><BR>> >> > Visit other Kitware open-source projects at<BR>> >> > http://www.kitware.com/opensource/opensource.html<BR>> >> ><BR>> >> > Please keep messages on-topic and check the ITK FAQ at:<BR>> >> > http://www.itk.org/Wiki/ITK_FAQ<BR>> >> ><BR>> >> > Follow this link to subscribe/unsubscribe:<BR>> >> > http://www.itk.org/mailman/listinfo/insight-users<BR>> >> ><BR>> >> ><BR>> ><BR>> > ________________________________<BR>> > Vive toda la experiencia de Messenger con Orange desde tu Blackberry<BR>> > ¡Descúbrelo!<BR>> > _____________________________________<BR>> > Powered by www.kitware.com<BR>> ><BR>> > Visit other Kitware open-source projects at<BR>> > http://www.kitware.com/opensource/opensource.html<BR>> ><BR>> > Please keep messages on-topic and check the ITK FAQ at:<BR>> > http://www.itk.org/Wiki/ITK_FAQ<BR>> ><BR>> > Follow this link to subscribe/unsubscribe:<BR>> > http://www.itk.org/mailman/listinfo/insight-users<BR>> ><BR>> ><BR><BR>
<HR>
¿Para qué descargarte juegos, si tienes los más divertidos online? <A href="http://juegosonline.es.msn.com/">Entra ya en Juegos y prepárate para muchas horas de diversión</A> <br /><hr />En tu material escolar no puede faltar el nuevo Pack de Emoticonos Vuelta al Cole <a href='http://www.vivelive.com/emoticonosvueltaalcole ' target='_new'>¡Descárgatelo gratis! Es muy divertido</a></body>
</html>