app.py

import fitz
import io
import base64
from PIL import Image
import gradio as gr
import cv2
import tempfile
import os

def pdf_to_img(pdf_path):
    pdf_document = fitz.open(pdf_path)
    counter = 1
    img_list = []
    for page_number in range(len(pdf_document)):
        page = pdf_document[page_number]
        image_list = page.get_images()
        for image in image_list:
            base_img = pdf_document.extract_image(image[0])
            image_data = base_img["image"]
            img = Image.open(io.BytesIO(image_data))
            # if img.mode == "RGBA":
            #     # Convert RGBA image to RGB
            #     img = img.convert("RGB")
            extention = base_img['ext']
            img.save(open(f"image{counter}.{extention}","wb"))
            img_list.append(f"image{counter}.{extention}")
            counter += 1
    return (img_list)

def extract_text_from_pdf(pdf_file):
    # Open the PDF file
    doc = fitz.open(pdf_file)
    
    # Initialize an empty string to store the extracted text
    extracted_text = ""
    
    # Iterate through each page of the PDF
    for page_num in range(len(doc)):
        # Load the page
        page = doc.load_page(page_num)
        
        # Extract text from the page and append it to the extracted_text string
        extracted_text += page.get_text()
    
    # Close the PDF document
    doc.close()
    return extracted_text


title = "Extract Image and Text"


with gr.Blocks(theme=gr.themes.Glass(primary_hue=gr.themes.colors.slate)) as demo:
    gr.Markdown(f'<h1 style="text-align: center;">{title}</h1>')
    with gr.Row():
        with gr.Row():
            with gr.Column():
                file_input = gr.File(type="filepath", label="Upload .pdf file")
                upload_button = gr.Button(value="Show Images")
                img_gallery = gr.Gallery(label="Generated images", show_label=True, elem_id="gallery", object_fit="contain", height="auto",allow_preview=True)
    
        with gr.Row():
            with gr.Column():
                output_text = text = gr.Textbox(label="Output", lines=4, autoscroll=False)
    upload_button.click(pdf_to_img, inputs=file_input, outputs=[img_gallery])
    upload_button.click(extract_text_from_pdf, inputs=file_input, outputs=[output_text])
demo.launch()