Julien, Many thanks for pointing this page to me. I understand documenting took a longtime and effort, therefore you should include the appropriate cross-references: in http://www.orfeo-toolbox.org/Applications/FineRegistration.html (which is the page referred to by otbcli_FineRegistration -help as "Complete documentation") and in http://www.orfeo-toolbox.org/CookBook/CookBooksu108.html
Note that the probability that a user reads the Software Guide from start to end is infinitesimal. The normal entry is through a tutorial (i.e. the Cook Book) and/or the application command help. And from there the user should get to the Software Guide if needed through the appropriate reference. Beyond this, **please note** that the metrics mentioned in http://www.orfeo-toolbox.org/Applications/FineRegistration.html "cross-correlation (CC), cross-correlation with subtracted mean (CCSM), mean-square difference (MSD), mean reciprocal square difference (MRSD) and mutual information (MI). Default is cross-correlation" **are not entirely coincident with those defined in the page you mention:** http://orfeo-toolbox.org/SoftwareGuide/SoftwareGuidech9.html#x32-1660009.7 which are: Mean squares Normalized correlation Mean reciprocal squared difference Mutual information by Viola and Wells Mutual information by Mattes Kullback Liebler distance metric by Kullback and Liebler Normalized mutual information Mean squares histogram Correlation coefficient histogram Cardinality Match metric Kappa Statistics metric Gradient Difference metric I understand the match for MSD and MRSD, but: 1. CC and CCSM are not defined in the Software Guide, at least not with the same names. 2. It is not clear which of the 3 mutual information indices is the one that is actually used Both problems leave the user confused. **But there is another problem: some of the definitions in the Software Guide do not seem correct** As this is another issue on its own, I open a different thread. Agus PS. we are still lacking a reasonable way in which the users could contribute to improve the documentation. Maybe a wiki would be a useful tool? Using git and alike is too complicated for the average user. On Thu, Nov 27, 2014 at 5:23 PM, Julien Michel <[email protected]> wrote: > Agustin, > > All those formulae were written in the ITK software guide, but I do not see > them anymore. I do not know why they removed them ... Luckily we have > imported them in our software guide (on which we actually spend a lot of > time ;) ): > > http://orfeo-toolbox.org/SoftwareGuide/SoftwareGuidech9.html#x32-1660009.7 > > Regards, > > Julien > > Le 27/11/2014 17:18, Agustin Lobo a écrit : > >> Julien, >> >> Both are pretty inextricable, >> I think that the formulae should be provided in the manual. >> Just one page with all formulae (which are not that many) >> and then each application can refer to that page if needed. >> I understand documentation is as hard as coding, but for serious use >> the package must be documented. We must know what we are doing >> and report it in the articles. >> >> Agus >> >> >> >> >> On Thu, Nov 27, 2014 at 3:13 PM, Julien Michel <[email protected]> >> wrote: >>> >>> Le 27/11/2014 15:06, Agustin Lobo a écrit : >>>> >>>> Are the metrics CC, NCC, CCSM, MRSD, MI available as buiilt-in in >>>> BandMath? >>> >>> No, but our new BandMath will allow for a lot of new features (already in >>> trunk, you can do a lot more). More info is coming on this soon. >>> >>>> And/Or could you provide (or point me to the appropriate doc for) the >>>> exact forms >>>> of these metrics that are used in otb? I'm finding differences between >>>> my own calculations >>> >>> See >>> >>> http://hg.orfeo-toolbox.org/OTB/file/8cbeb047c346/Applications/DisparityMap/otbFineRegistration.cxx >>> >>> All metrics are actually implemented in ITK, so you can refer to their >>> documentation, for instance : >>> >>> >>> http://www.itk.org/Insight/Doxygen/html/classitk_1_1MattesMutualInformationImageToImageMetric.html >>> >>> Regards, >>> >>> Julien >>> >>> -- >>> Julien MICHEL >>> CNES - DCT/SI/AP - BPI 1219 >>> 18, avenue Edouard Belin >>> 31401 Toulouse Cedex 09 - France >>> Tel: +33 561 282 894 - 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. > > > > -- > Julien MICHEL > CNES - DCT/SI/AP - BPI 1219 > 18, avenue Edouard Belin > 31401 Toulouse Cedex 09 - France > Tel: +33 561 282 894 - 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. -- -- 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.
