Update app.py

app.py

@@ -6,6 +6,7 @@ from transformers import TrOCRProcessor, VisionEncoderDecoderModel
 import gradio as gr
 import numpy as np
 from PIL import Image
+import io
 
 # Load models on CPU (Hugging Face Spaces default)
 craft_model = Model.load("hezarai/CRAFT", device="cpu")
@@ -22,6 +23,10 @@ def recognize_handwritten_text(image):
         # Load image with hezar utils using file path
         processed_image = load_image(tmp_path)
         
+        # Ensure processed_image is in a compatible format (convert to NumPy if needed)
+        if not isinstance(processed_image, np.ndarray):
+            processed_image = np.array(Image.open(tmp_path))
+        
         # Detect text regions with CRAFT
         outputs = craft_model.predict(processed_image)
         if not outputs or "boxes" not in outputs[0]: