How to Redact Documents (PDF, JPG, PNG) using Python

What is Document Redaction?

Document redaction, also known as anonymization, is the process of hiding or removing sensitive information, such as names, dates, or identification numbers from documents.

This helps protect privacy and ensures compliance with data protection regulations by making confidential data unreadable or invisible.

‍

How to Redact Documents

‍

Set Up Your Eden AI Account

1. Create an account: If you don't have an Eden AI account, create one for free using the following link. Then you can obtain your API key in the API Keys section, which you can use to access various AI services.

2. Access OCR Tools: Once logged in, go to the document parsing section of the platform.

3. Choose the Document Redaction Feature: Select the Document Redaction tool. You can also explore other OCR tools.

‍

Implementing Document Redaction in Python

The process is asynchronous to allow for better performance and scalability. Redacting a document can take time depending on file size and complexity, so:

• The POST request initiates the job and returns immediately.
• The GET request lets you check when the job is complete and fetch the result.

This two-step system is ideal for batch processing or workflows where real-time response isn’t required.

‍

pip install requests

Step 2: Launch a Redaction Job (POST Request)

The first step is to initiate the redaction process by launching a job. You’ll need to send the image URL to the API for processing.

import requests

headers = {"Authorization": "Bearer {API_TOKEN}"}
url = "https://api.edenai.run/v2/ocr/anonymization_async"
json_payload = {
    "providers": "readyredact",
    "file_url": "🔗 URL of your image"
}

response = requests.post(url, json=json_payload, headers=headers)

result = response.json()
print(result)

{
  "public_id": "1234567890abcdef",
  "status": "started"
}

Step 3: Get the Redaction Results (GET Request)

Once the job is launched, you need to fetch the results using the public_id.

‍

import requests

headers = {
    "Authorization": "Bearer {API_TOKEN}"
}
url = f'https://api.edenai.run/v2/ocr/anonymization_async/{public_id}/'

response = requests.get(url, headers=headers)

result = response.json()
print(result)

{
  "status": "finished",
  "results": {
    "readyredact": {
      "status": "success",
      "redacted_document_url": "https://url_to_redacted_file.pdf"
    }
  }
}

Managing Redaction Jobs

In addition to launching and retrieving results from redaction jobs, Eden AI provides optional endpoints to help you manage your anonymization tasks more efficiently:

1. List All Redaction Jobs
   Endpoint: GET https://api.edenai.run/v2/ocr/anonymization_async/
   Use this to retrieve a list of all launched redaction jobs. It’s helpful for tracking job statuses or fetching results later using their job IDs.
   Note: A job's status only updates after a GET request is made. You can check the documentation here.
   
   
2. Delete Redaction Jobs
   Endpoint: DELETE https://api.edenai.run/v2/ocr/anonymization_async/
   This allows you to delete redaction jobs from your account once you’ve retrieved the results, helping you keep your job history clean and organized.
   You can check the documentation here.

‍

Why Eden AI is the Best Tool for Document Redaction

Eden AI provides several advantages.

Conclusion

With just a few lines of code, you can automate document redaction using Python and Eden AI.

The two-step process ensures accuracy and flexibility, especially for larger workloads.

Whether you’re processing legal files, patient records, or financial data, Eden AI provides a scalable and secure solution for anonymization.

Start redacting smarter, not harder, with Eden AI.

You can access the full documentation for Document Redaction here.