Create app.py
Browse files
app.py
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import gradio as gr
|
| 2 |
+
import pytesseract
|
| 3 |
+
from pdf2image import convert_from_path
|
| 4 |
+
import os
|
| 5 |
+
|
| 6 |
+
# Function to perform OCR and search for a keyword
|
| 7 |
+
def ocr_pdf_and_search_keyword(pdf_file, keyword, lang='fas'): # 'fas': Persian language (Farsi)
|
| 8 |
+
# Convert PDF to images
|
| 9 |
+
images = convert_from_path(pdf_file.name)
|
| 10 |
+
|
| 11 |
+
# Initialize a variable to hold the extracted text
|
| 12 |
+
extracted_text = ""
|
| 13 |
+
keyword_found = False
|
| 14 |
+
|
| 15 |
+
# Loop through each image and perform OCR
|
| 16 |
+
for page_number, image in enumerate(images):
|
| 17 |
+
text = pytesseract.image_to_string(image, lang=lang)
|
| 18 |
+
extracted_text += text
|
| 19 |
+
|
| 20 |
+
# Check if the keyword is in the extracted text
|
| 21 |
+
if keyword.lower() in text.lower():
|
| 22 |
+
keyword_found = True
|
| 23 |
+
|
| 24 |
+
if not keyword_found:
|
| 25 |
+
result_message = f"Keyword '{keyword}' not found in the document."
|
| 26 |
+
else:
|
| 27 |
+
result_message = f"Keyword '{keyword}' found in the document."
|
| 28 |
+
|
| 29 |
+
return extracted_text, result_message
|
| 30 |
+
|
| 31 |
+
# Create Gradio interface
|
| 32 |
+
def gradio_interface():
|
| 33 |
+
# Define Gradio inputs and outputs
|
| 34 |
+
pdf_input = gr.File(label="Upload PDF")
|
| 35 |
+
keyword_input = gr.Textbox(label="Enter Keyword", value="فلسفه") # Default keyword is 'فلسفه'
|
| 36 |
+
output_text = gr.Textbox(label="Extracted Text", interactive=False)
|
| 37 |
+
output_message = gr.Textbox(label="Keyword Search Result", interactive=False)
|
| 38 |
+
|
| 39 |
+
# Function to process the inputs and return the outputs
|
| 40 |
+
def process(pdf_file, keyword):
|
| 41 |
+
extracted_text, result_message = ocr_pdf_and_search_keyword(pdf_file, keyword)
|
| 42 |
+
return extracted_text, result_message
|
| 43 |
+
|
| 44 |
+
# Create and launch Gradio interface
|
| 45 |
+
gr.Interface(fn=process,
|
| 46 |
+
inputs=[pdf_input, keyword_input],
|
| 47 |
+
outputs=[output_text, output_message],
|
| 48 |
+
title="OCR PDF Keyword Search",
|
| 49 |
+
description="Upload a PDF, enter a keyword, and see the OCR results along with a search for the keyword."
|
| 50 |
+
).launch()
|
| 51 |
+
|
| 52 |
+
# Call the function to create the interface
|
| 53 |
+
gradio_interface()
|