OCR Passport Extractor

The Machine Readable Zone is the two-line code at the bottom of a passport page. It stores standardized data and gives the most reliable surname and given-name extraction.

Yes. All OCR runs locally in your browser via Tesseract.js. The image is never uploaded, which matters for identity documents.

The MRZ lines are parsed for surname and given names, while passport number, nationality, and date of birth come from pattern matches over the full OCR text.

The number pattern expects two letters followed by six to nine digits. Cropped, rotated, or low-contrast scans often fail; retry with a clearer image.

The raw OCR text is shown so you can see what the engine actually read. Improving lighting, focus, and frame coverage usually fixes detection.

Yes. The same language selector used across the OCR tools is available here, defaulting to English, and additional bundled languages can improve recognition for non-English passports.