File size: 1,464 Bytes
284e9bf 7ad7c0a 284e9bf 7ad7c0a 875e5d4 284e9bf 7ad7c0a 284e9bf 7a1ecf6 875e5d4 284e9bf 5ea229f 7a1ecf6 284e9bf 5ea229f 284e9bf 5ea229f |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 |
import pytesseract
from pdf2image import convert_from_path
import google.generativeai as genai
import os, json
# --- Form Schema ---
FORMS = {
"pancard_form": [
"Name", "DOB", "Gender", "FatherName", "MotherName",
"Address", "City", "State", "Pincode", "Mobile", "Email",
"DocumentType", "DocumentNumber", "IssueAuthority",
"IssueDate", "ExpiryDate"
]
}
# --- Configure Gemini ---
api_key = os.getenv("GEMINI_API_KEY")
if not api_key:
raise ValueError("❌ GEMINI_API_KEY not found. Please set it in Hugging Face Space Secrets.")
genai.configure(api_key=api_key)
def extract_text_from_pdf(pdf_path):
pages = convert_from_path(pdf_path)
text = ""
for page in pages:
text += pytesseract.image_to_string(page) + "\n"
return text.strip()
def extract_key_values_with_gemini(raw_text, form_type="pancard_form"):
prompt = f"""
You are an intelligent document parser.
Given the following document text, extract only these fields: {FORMS[form_type]}.
Return strictly as JSON key-value pairs.
Document text:
{raw_text}
"""
model = genai.GenerativeModel("models/gemini-2.5-flash")
print("Gemini API called successfully ✅")
response = model.generate_content(prompt)
text = response.text.strip()
# --- Cleanup ---
text = text.replace("```json", "").replace("```", "").strip()
try:
return json.loads(text)
except Exception:
return {"raw_output": text} |