Community

Character position

Hello Help, how can i get the original Position of a char after i deskew a picture for recognition?

UPD:

now i have solved it.

i use the function convertCoordinate. There i create 2 imagedoc object. The one i use it for Recognition... and the other one i use it for convertcoordinate. If you use a only a region on a imagedoc over imagemodification, there you cant use the convertcoordinate. That the reason that i have create 2 imagedoc.

0

Comments

3 comments

  • Avatar
    Dmitry Mesheryakov

    This should be a separate question and you should provide more details.

    0
    Comment actions Permalink
  • Avatar
    Anastasia Galimova

    To get the original position of a character after you deskew a picture for recognition you can use the following algorithm:

    1. Get the position of a char on the deskewed image using the CharacterRegion object (see Help > Index > CharacterRegion).
    2. Use ConvertCoordinates method to convert the coordinates on the deskewed image to the coordinates on the original image (see the corresponding sample in our Knowledge Base).

    I can also provide a sample code if you specify the programming language you use.

    0
    Comment actions Permalink
  • Avatar
    patrick

    Hi Anastasia,

    if you could post a C# sample that would be great - for some reason the link did not work for me.

    We are using the FR10 SDK and exporting the output to XML BTW - is there a way to just get the affine transformation (shift/rotation applied) included in the XML file?

    Regards,

    Patrick

    0
    Comment actions Permalink

Please sign in to leave a comment.