Hi Eric, I did not look in detail to the implementation but I know that textures computation in this filter is mostly based on the implementation available in ITK class HistogramToTextureFeaturesFilter:
http://www.itk.org/Doxygen/html/classitk_1_1Statistics_1_1HistogramToTextureFeaturesFilter.html Please let us know if you find something wrong between formulas provided in ITK and OTB dowygen documentation and references and the current implementation Note also that there is a test in OTB which create "by hand" a co-occurence matrix and compute texture features (could help to investigate possible problems): http://hg.orfeo-toolbox.org/OTB/file/b7cbca69fc3d/Testing/Code/FeatureExtraction/otbGreyLevelCooccurrenceMatrixAdvancedTextureCoefficientsCalculator.cxx Regards, Manuel Le 22/09/2014 14:53, Erik a écrit : > Hello Everybody! > > i have a question concerning the derivation of Haralick features > implemented in the > | > > > otb::ScalarImageToTexturesFilter > > | > class: > > Perhaps i am missing something but i do not understand why a different > mean value is used for the Haralicks Correlation feature and the rest > of the features. Following this i do not get what the > | > marginalSum > | > variable represents and why it is used to calculate the mean and > deviation for the Haralick Correlation feature. > I am referring to the file /otbHaralickTexturesImageFunction.txx/ > (doc: > http://orfeo-toolbox.org/doxygen/otbHaralickTexturesImageFunction_8txx_source.html) > > Below is the part of the code my question is related to > | > while(it !=glcVector.end()) > { > doublefrequency =(*it).second /totalFrequency; > CooccurrenceIndexType > <http://orfeo-toolbox.org/doxygen/classitk_1_1Index.html>index > =(*it).first; > pixelMean +=index[0]*frequency; > marginalSums[index[0]]+=frequency; > ++it; > } > -------------------------------------------------- > textures[7]+=index[0]*index[1]*frequency; > textures[7]=(fabs(marginalDevSquared)>1E-8)?(textures[7]-marginalMean > *marginalMean )/marginalDevSquared :0; > > | > > Many thanks in advance for your help! > Greetings, > Erik > -- > -- > Check the OTB FAQ at > http://www.orfeo-toolbox.org/FAQ.html > > You received this message because you are subscribed to the Google > Groups "otb-users" group. > To post to this group, send email to [email protected] > To unsubscribe from this group, send email to > [email protected] > For more options, visit this group at > http://groups.google.com/group/otb-users?hl=en > --- > You received this message because you are subscribed to the Google > Groups "otb-users" group. > To unsubscribe from this group and stop receiving emails from it, send > an email to [email protected] > <mailto:[email protected]>. > For more options, visit https://groups.google.com/d/optout. -- Manuel GRIZONNET CNES - DCT/SI/AP - BPI 1219 18, avenue Edouard Belin 31401 Toulouse Cedex 09 - France Tel: +33 561 282 630 - Fax: +33 561 283 109 -- -- Check the OTB FAQ at http://www.orfeo-toolbox.org/FAQ.html You received this message because you are subscribed to the Google Groups "otb-users" group. To post to this group, send email to [email protected] To unsubscribe from this group, send email to [email protected] For more options, visit this group at http://groups.google.com/group/otb-users?hl=en --- You received this message because you are subscribed to the Google Groups "otb-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
