Thank you for the link, Zdenko. Since I also want to get the features from ready-to-use traineddata files, that don't come with box/tr files, I started using the methods in tessdatamanager.h and intprotos.h.
Do you know how the features in tr files differ from the data in the intproto section of a traineddata file? I know that intproto mainly contains the 4D (x, y, angle, length) information. There are four sections for each box in a tr file: mf, cn, if and tb. if seem to be the 3D int features (x, y, angle), but what do the features at mf, cn and tb mean? Paul Am Montag, 7. Juli 2014 18:38:25 UTC+2 schrieb zdenop: > > Maybe this will help you[1]. > [1] > https://groups.google.com/forum/?utm_medium=email#!msg/tesseract-dev/EYOfbGXiiIc/JtjB8q4lMjcJ > > Zdenko > > > On Mon, Jul 7, 2014 at 3:00 PM, Paul <[email protected] <javascript:>> wrote: > >> Is there a method in the public API that allows me to gather the >> information from normproto or inttemp? >> >> Am Samstag, 5. Juli 2014 16:42:30 UTC+2 schrieb Paul: >> >>> I found out the following: >>> >>> - int_features should be an array of length 512, >>> - num_features tells us how many features are really in int_features >>> - FeatureOutlineIndex holds the index of the corresponding outline >>> for each feature in int_features >>> >>> TESS_API TBLOB* >>> TESS_CALL TessMakeTBLOB(Pix* pix); >>> >>> will give us the TBLOB* needed for TessBaseAPIGetFeaturesForBlob(). If >>> you want to get the Pix* for every symbol, simply call >>> TessPageIteratorGetImage() or TessPageIteratorGetBinaryImage() while >>> iterating over the symbols. >>> >>> Now I still need to know how to get the features for prototypes... >>> >>> >>> Am Samstag, 5. Juli 2014 14:37:20 UTC+2 schrieb Paul: >>>> >>>> Hello everyone, >>>> >>>> I am trying to get a symbol's features by using the C API so I can >>>> create additional ways for debugging Tesseract. >>>> >>>> Can anybody provide me with an example on how to use the following >>>> method? >>>> >>>> TESS_API void TESS_CALL TessBaseAPIGetFeaturesForBlob(TessBaseAPI* >>>> handle, TBLOB* blob, INT_FEATURE_STRUCT* int_features, >>>> int* >>>> num_features, int* FeatureOutlineIndex); >>>> >>>> As far as I understand it, I need to provide an INT_FEATURE_STRUCT that >>>> will hold the resulting features of the symbol after calling the method, >>>> as >>>> well as an int*, which will hold the number of features. What goes in >>>> FeatureOutlineIndex and which TBLOB* do I have to provide? How to I >>>> get it? >>>> >>>> Or is there another method that can give me the features of a symbol? >>>> >>>> Best regards, >>>> Paul >>>> >>> -- >> You received this message because you are subscribed to the Google Groups >> "tesseract-ocr" group. >> To unsubscribe from this group and stop receiving emails from it, send an >> email to [email protected] <javascript:>. >> To post to this group, send email to [email protected] >> <javascript:>. >> Visit this group at http://groups.google.com/group/tesseract-ocr. >> To view this discussion on the web visit >> https://groups.google.com/d/msgid/tesseract-ocr/6b0f5ab3-4c56-4236-aa61-8ae99165cb8c%40googlegroups.com >> >> <https://groups.google.com/d/msgid/tesseract-ocr/6b0f5ab3-4c56-4236-aa61-8ae99165cb8c%40googlegroups.com?utm_medium=email&utm_source=footer> >> . >> >> For more options, visit https://groups.google.com/d/optout. >> > > -- You received this message because you are subscribed to the Google Groups "tesseract-ocr" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/tesseract-ocr. To view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/fbef68ce-3b65-4ff7-b0db-161e62c78c30%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
