zafranf / ktp-ocr

An Open Source OCR tool for Indonesian ID card (KTP).

KTP-OCR

Kartu Tanda Penduduk Extractor
An attempt to create a production grade KTP extractor.

KTP-OCR is a open source python package that attempts to create a production grade KTP extractor. The aim of the package is to extract as much information as possible yet retain the integrity of the information.

---

Requirements

You will need tesseract with indonesian language support installed in your system.

$ brew install tesseract-lang

---

🚀 How to launch

$ git clone https://github.com/YukaLangbuana/KTP-OCR.git
$ cd KTP-OCR
$ pip install -r requirements.txt
$ python3 ocr.py <path-image>

---

📝 Note from Yuka

• I am actively working to create a python package out of the main ocr.py. For now you can play with the old script.
• I have an idea to verify the address information from the KTP via external service (Google Maps) which can be used to further standardized Indonesian address' information.