add redact

app.py

@@ -5,7 +5,12 @@ import gradio as gr
 from mistralai import Mistral, TextChunk, ImageURLChunk
 from PIL import Image
 import requests
+from presidio_image_redactor import ImageRedactorEngine
 
+def redact_imag(img):
+    redactor = ImageRedactorEngine()
+    redact_img = redactor.redact(image=img, entities=["PERSON", "LOCATION", "DATE_TIME", "PHONE_NUMBER","MEDICAL_LICENSE"])
+    return redact_img
 
 def process_msg(client, target):
 
@@ -113,9 +118,10 @@ with gr.Blocks(
 
             submit_btn = gr.Button("Extract Text and Images")
 
-        with gr.Column(scale=2):
+        with gr.Column(scale=3):
+            image_output = gr.Gallery(label="Redacted Image", height="contain")
             ocr_result = gr.Textbox(label="Annoymized Text", show_copy_button=True)
-            cleaned_output = gr.Textbox(label="Extracted Plain Text", lines=10, show_copy_button=True)
+            cleaned_output = gr.Textbox(label="Extracted JSON", lines=10, show_copy_button=True)
 
     def update_visibility(choice):
         return gr.update(visible=(choice == "URL")), gr.update(visible=(choice == "Upload file"))

packages.txt

@@ -0,0 +1 @@
+tesseract-ocr-all

requirements.txt

@@ -1,6 +1,3 @@
 mistralai
-pillow
-fastai
-transformers
-paddlepaddle
-paddleocr
+presidio-image-redactor
+pytesseract