Skip to content

From identity card image, this repo detect 4 corners, align by OpenCV, then detect word in image and recognize word by Transformer OCR.

Notifications You must be signed in to change notification settings

chamnab-sonn/extract-information-from-identity-card

This branch is 2 commits behind buiquangmanhhp1999/extract-information-from-identity-card:master.

Repository files navigation

Extract-information-from-identity-card

Tutorial: Recognize Identity Card Tutorial

1. Pipeline

1.1. Corner Detection and Alignment

1.2. Text Detection

1.3. Recognize image text and final results

Final Result

{
  "id": "38138424",
  "name": "LÊ KIỀU DIỄM",
  "birth": "1989",
  "add": "Tân Hưng Tây Phú Tân Cà Mau",
  "home": "Khóm 8 Phường 8 TP Cà Mau Cà Mau"
}

2. How to run project

2.1. Install dependecies

pip install -r requirement.txt

2.2. Download weight from gg drive

Download weight from here , then put into src/vietocr/config_text_recognition folder.

2.3. Start server

python main.py

About

From identity card image, this repo detect 4 corners, align by OpenCV, then detect word in image and recognize word by Transformer OCR.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 100.0%