Update app.py

app.py

@@ -18,27 +18,20 @@ def get_image_embeddings(image_path, model_name='google/vit-base-patch16-224'):
  return embeddings
 
 # Function to convert PDF to images
-def pdf_to_images(pdf_file, img_dir):
- # Open the provided PDF file
- doc = fitz.open(pdf_file)
+from pdf2image import convert_from_path
 
+def pdf_to_images(pdf_file, img_dir):
+ images = convert_from_path(pdf_file)
+ 
  # Create the directory if it doesn't exist
  os.makedirs(img_dir, exist_ok=True)
 
- for page_num in range(len(doc)):
- # Get the page
- page = doc.load_page(page_num)
-
- # Render the page to an image
- pix = page.get_pixmap()
-
- # Define the output image path
- output_file = f"{img_dir}/page_{page_num + 1}.png"
+ for i, image in enumerate(images):
+ image_path = f"{img_dir}/page_{i + 1}.png"
+ image.save(image_path, "PNG")
 
- # Save the image
- pix.save(output_file)
+ print(f"Converted {len(images)} pages to images and saved in {img_dir}")
 
- print(f"Converted {len(doc)} pages to images and saved in {img_dir}")
 
 # Function to get text embeddings using a transformer model
 def get_text_embeddings(text, model_name='bert-base-uncased'):