You could use OpenCV to define a template with regions of interest (ROIs) and then use tesseract to OCR them?
On Tuesday, January 25, 2022 at 3:52:57 AM UTC-7 [email protected] wrote: > I'm also facing the same issue..... Can someone throw some time light here > please > > > > On Saturday, January 11, 2020 at 12:51:09 PM UTC+5:30 > [email protected] wrote: > >> yes. that's what I am doing right now. but there's an issue. >> >> While I am getting all the names, I am not able to detect the [father's >> name] even if it is in the lot and some unwanted garbage is sometimes >> getting in the way. >> >> I am attaching a sample image for your reference. and the text generated >> from that (in a list). >> >> >> ['*MONIKA MAHADEV SHINDE*', '*ARa Gar*', 'GOVT. OF INDIA', '*MAHADEV >> SHINDE*', '*31/10/1992*', 'Permanent Account Number', '*EJAPS0276M *~', >> 'MONIKA 1 SHIN OE :', '- 8', 'Signature'] >> >> >> I am obtaining the above list. and the father's name shows "ARa Gar" and >> not MAHADEV SHINDE. >> As the unwanted text is not generated always, I need a way to figure out >> what the actual name might be. >> >> Can you please look into this?? >> >> >> On Wednesday, 8 January 2020 01:02:56 UTC+5:30, Saurabh Pal wrote: >>> >>> Try using template matching for your use case(I am assuming that PAN >>> card format is same all over India). Atleast dob and pan card number can be >>> found easily using regex. For names, you can reject all the other text >>> boxes like 'INCOME TAX DEPARTMENT', 'GOVT. OF INDIA', then you will be left >>> with only the fathers name and holders name just check for the y coordinate >>> among those two text boxes. >>> >>> >>> On Monday, January 6, 2020 at 7:37:13 PM UTC+5:30, Shubhranshu Panda >>> wrote: >>>> >>>> I don't know how to extract particular text from a standard image type. >>>> I want to extract the name, dob and PAN number from the PAN card. I have >>>> attachhed a sample image for reference. >>>> >>> -- 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 view this discussion on the web visit https://groups.google.com/d/msgid/tesseract-ocr/0176ded0-d9d3-4b23-b102-60a39e1ccc92n%40googlegroups.com.
