Trying to extract text from scanned paper forms containing logos, lines, boxes, and text. From what I've read, I expected Tesseract to segment the page and classify each element. I tried TessBaseAPI::SetPageSegMode () with PSM_AUTO, PSM_AUTO_OSD and a few others, followed by TessBaseAPI::AnalyseLayout (), but all I get is a single PT_FLOWING_IMAGE block representing the whole page.
However, if I REMOVE the logo from the form, I then get PT_FLOWING_TEXT, PT_HORZ_LINE, and PT_VERT_LINE blocks, and TessBaseAPI::Recognize does a fairly good job recognizing the text, even though the text is not in contiguous blocks and is interspersed among the lines. I have seen examples online of Tesseract segmenting a page and separately identifying blocks of text and graphics, but I cannot remember where. So, I am looking for information and advice on how to have Tesseract accurately segment a form that includes images and accurately recognize text interspersed among lines and boxes. Thank you. -- 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/63cb5a8a-3a5b-478b-a73a-8b946367b8b0%40googlegroups.com. For more options, visit https://groups.google.com/d/optout.
