Update main.py

main.py

@@ -1,33 +1,29 @@
-# For Google Colab
-from google.colab import files
-from PIL import Image
+from fastapi import FastAPI
+import base64
+from PIL import Image, ImageEnhance
 import pytesseract
 from langdetect import detect, DetectorFactory
-from googletrans import Translator
+from deep_translator import GoogleTranslator
 import re
-import io
-from pprint import pprint
 import numpy as np
 import cv2
-from PIL import Image, ImageEnhance, ImageFilter
 import unicodedata
+import io
+from pydantic import BaseModel
+
+pytesseract.pytesseract.tesseract_cmd = "/usr/bin/tesseract"
 
 # Fix language detection randomness
 DetectorFactory.seed = 0
 
-def upload_and_process_image():
-    print(" Please upload an invoice image:")
-    uploaded = files.upload()
-    if not uploaded:
-        print("No file uploaded.")
-        return None
-    image_path = list(uploaded.keys())[0]
-    try:
-        image = Image.open(image_path)
-        return image
-    except Exception as e:
-        print("Error loading image:", e)
-        return None
+app = FastAPI()
+
+LANG_CODE_MAP = {
+    "en": "eng", "ta": "tam", "hi": "hin",
+    "kn": "kan", "ml": "mal", "te": "tel",
+    "bn": "ben", "gu": "guj", "pa": "pan", "mr": "mar"
+}
+
 
 def perform_ocr(image):
     try:
@@ -73,17 +69,6 @@ def perform_ocr(image):
         return None
 
 
-LANG_CODE_MAP = {
-    "en": "eng",
-    "ta": "tam",
-    "hi": "hin",
-    "kn": "kan",
-    "ml": "mal",
-    "te": "tel",
-}
-
-
-
 def clean_ocr_text(text):
     # Normalize unicode (fix weird diacritics, spacing issues)
     text = unicodedata.normalize("NFKC", text)
@@ -114,14 +99,6 @@ def clean_ocr_text(text):
     return text
 
 
-def preprocess_image(image):
-    """Convert to grayscale, remove noise, and improve text clarity."""
-    gray = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
-    gray = cv2.threshold(gray, 150, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)[1]
-    gray = cv2.medianBlur(gray, 3)
-    return gray
-
-
 def preprocess_image(image):
     if image is None: # Check if image is None
         print("Error: Input image is None.")
@@ -166,20 +143,26 @@ def detect_language(text_data):
         return None
 
 
-def translate_text(text_data, src_lang):
-    """Translate text to English if needed"""
-    if src_lang == 'en':
-        print("\nText is already in English, no translation needed.")
-        return text_data['original_text']
+def perform_ocr(image):
+    text = pytesseract.image_to_string(
+        image,
+        lang='eng+tam+kan+hin+tel+mal+ben+guj+pan+mar',
+        config='--psm 6'
+    ).strip()
+    detected_lang = detect(text) if text else "en"
+    translated_text = None
+    if detected_lang != 'en' and text:
+        try:
+            translated_text = GoogleTranslator(source=detected_lang, target="en").translate(text)
+        except Exception as e:
+            translated_text = f"[Translation failed: {e}]"
+
+    return {
+        "detected_language": detected_lang,
+        "original_text": text,
+        "translated_text": translated_text
+    }
 
-    try:
-        translator = Translator()
-        translation = translator.translate(text_data['original_text'], src=src_lang, dest='en')
-        print("\nTranslation to English completed.")
-        return translation.text
-    except Exception as e:
-        print(f"Translation error: {e}")
-        return text_data['original_text']
 
 
 def extract_field_from_lines(lines, patterns):