Update app.py

app.py

@@ -1,31 +1,26 @@
 import gradio as gr
+from transformers import TrOCRProcessor, VisionEncoderDecoderModel
 from PIL import Image
-import pytesseract
-import cv2
-import numpy as np
+
+# Load the TrOCR model for printed text
+processor = TrOCRProcessor.from_pretrained("microsoft/trocr-base-printed")
+model = VisionEncoderDecoderModel.from_pretrained("microsoft/trocr-base-printed")
 
 def ocr_extract(image):
     if image is None:
-        return "Please upload an image file."
-    
-    # Convert PIL image to OpenCV format
-    image_np = np.array(image)
-    image_cv = cv2.cvtColor(image_np, cv2.COLOR_RGB2BGR)
-
-    # (Optional) Preprocess for better accuracy
-    gray = cv2.cvtColor(image_cv, cv2.COLOR_BGR2GRAY)
-    _, thresh = cv2.threshold(gray, 150, 255, cv2.THRESH_BINARY)
-
-    # OCR
-    text = pytesseract.image_to_string(thresh)
-    return text.strip()
+        return "Please upload an image."
+    image = image.convert("RGB")
+    pixel_values = processor(images=image, return_tensors="pt").pixel_values
+    generated_ids = model.generate(pixel_values)
+    extracted_text = processor.batch_decode(generated_ids, skip_special_tokens=True)[0]
+    return extracted_text.strip()
 
 iface = gr.Interface(
     fn=ocr_extract,
     inputs=gr.Image(type="pil"),
     outputs="text",
-    title="Bill Text Scanner (Tesseract)",
-    description="Upload a scanned bill or screenshot to extract all readable text."
+    title="Bill OCR Scanner (Printed Text)",
+    description="Upload a machine-printed bill or receipt to extract text."
 )
 
 if __name__ == "__main__":