backend/text_recog.py

import pytesseract
import cv2
import re
import platform
from .file_utils import convert_image_to_word

def configure_tesseract():
    system = platform.system()
    if system == "Windows":
        pytesseract.pytesseract.tesseract_cmd = r"C:\Users\hp\AppData\Local\Programs\Tesseract-OCR\tesseract.exe"
    else:
        pytesseract.pytesseract.tesseract_cmd = "/usr/bin/tesseract"

def extract_ktp_info(image_path, filename):
    configure_tesseract()
    
    img = cv2.imread(image_path)
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    text = pytesseract.image_to_string(gray, lang='ind')

    text = text.replace('\n\n\n', '\n').replace('\n\n', '\n').replace('\n', '\n')

    match = re.search(r'Berlaku Hingga\n(.*?)\nGol\. Darah', text, re.DOTALL)

    if match:
        extracted_text = match.group(1)
        lines = [
            re.sub(r'^(:|\d+)?\s*', '', line.strip()) 
            for line in extracted_text.strip().split('\n')
            if line.strip()
        ]
        print("Hasil List Bersih:")
        print(lines)
    
    city = re.search(r'PROVINSI\s+(.+?)\n(.+?)\n', text)

    result = {
        "nik" : lines[0],
        "nama" : lines[1],
        "tempat_tgl_lahir" : lines[2],
        "jenis_kelamin" : lines[3],
        "alamat" : lines[4],
        "rt_rw" : lines[5],
        "kel/desa" : lines[6],
        "kecamatan" : lines[7],
        "provinsi" : lines[13],
        "agama" : lines[8],
        "kewarganegaraan" : lines[10],
        "pekerjaan" : lines[9],
    }

    # convert_image_to_word(result, f'KTP {filename}')

    return result