Extract the passport holder’s name directly from a scanned passport image

A few days ago, I stumbled upon a fun little challenge: extract the passport holder’s name directly from a scanned passport image. At first glance, it seemed simple: just run OCR and grab the text. But if you’ve ever worked with OCR, you know the output can get messy.

So I rolled up my sleeves and built a small Python function: parse_passport_holder_name.
It uses Tesseract OCR and a bit of regex magic to detect the MRZ pattern (the machine-readable zone at the bottom of a passport) and automatically extracts the person’s name.

When the MRZ is clean, the regex works perfectly. And when it’s not, I added a fallback that guesses possible names from the OCR text.

https://gist.github.com/amirshnll/66aae6fd07b6c8f2c41521253487152d