Figure, Graph, Image detection/classification using Tesseract OCR

[Mohit Jain]

I'd like to know if it's possible to use Tesseract OCR for automatically detecting figures, graphs or images which occur in the image? From reviewing the code-documentation, I can see that it's possible to expose the AnalyseLayout class by minor changes to the C++ API. I would like to know if Tesseract performs any image/graph/figure detection (exposed by the API or otherwise - like AnalyseLayout) to improve its recognition and layout results? Thanks!

[nick]

this is my question too, Does anyone have an idea for this ?