Feature: Integrate Tesseract as the ultimate tertiary OCR engine for maximum text extraction power

packages.txt

@@ -1,2 +1,5 @@
 libgl1
 libglib2.0-0
+tesseract-ocr
+tesseract-ocr-eng
+tesseract-ocr-tam

pipeline/ocr.py

@@ -89,8 +89,23 @@ def extract_text_from_image(image_path: str) -> str:
                         all_text.extend(result)
                     except Exception as e:
                         print(f"DEBUG: EasyOCR Failed: {e}")
+                        
+                        try:
+                            # --- TERTIARY: Tesseract OCR (Ultimate Open-Source Backup) ---
+                            print("DEBUG: Falling back to Tesseract AI...")
+                            import pytesseract
+                            from PIL import Image
+                            # Tesseract usually performs best with PIL images
+                            pil_img = Image.fromarray(cv2.cvtColor(thresh, cv2.COLOR_GRAY2RGB))
+                            # psm 6 assumes a single uniform block of text
+                            tess_text = pytesseract.image_to_string(pil_img, lang='eng', config='--psm 6')
+                            if tess_text.strip():
+                                all_text.append(tess_text.strip())
+                        except Exception as e:
+                            print(f"DEBUG: Tesseract Failed (Ensure tesseract is installed on system): {e}")
+                            
         else:
-            # Use EasyOCR Fallback
+            # Use EasyOCR Fallback if Paddle isn't available
             reader = get_easy_reader()
             if not reader: return "OCR Engine Error"
             

requirements.txt

@@ -14,3 +14,4 @@ deep-translator
 psutil
 opencv-python
 nest_asyncio
+pytesseract