[Python] How to Integrate an Image Classifier with OpenDLP for Sensitive and Non-Sensitive...

I'm working on an open-source Data Loss Prevention (DLP) solution trying to integrate image scanning capabilities using machine learning (ML) for image classification. I’ve set up a reverse proxy with two endpoints: one for OpenDLP (running on a legacy Ubuntu VM) and another for my image classifier (running on the host OS).

However, I’m facing difficulties integrating the results from the image classifier (which classifies images as sensitive or non-sensitive) into OpenDLP, and I need help with how to approach this integration.

Setup:
OpenDLP is running in a VM on an older version of Ubuntu (2.4) using the LAMP stack. OpenDLP has not been updated in over a decade. Image Classifier is running on the host OS. It uses ML for classifying images as sensitive or non-sensitive. Reverse Proxy: I’ve set up a reverse proxy with two endpoints: One for OpenDLP. Another for Image Scanning/Classification (using the image classifier).

The Challenge:
OpenDLP Integration: I need to integrate the classification results (sensitive or non-sensitive) from my image classifier into the OpenDLP system. The classifier is working fine, but I don't know how to forward its results (in a structured format) to OpenDLP.

Reverse Proxy: Since OpenDLP is running in a VM with outdated services (Python2, old versions of libraries), upgrading it is quite painful. This is why I’m opting for a reverse proxy setup to keep OpenDLP in the VM and run the classifier on the host OS.

API Communication: I’m not familiar with how to make API calls between a LAMP stack (OpenDLP) and a Python-based image classifier. How can I integrate the results from the classifier into OpenDLP without modifying OpenDLP’s source code directly?

What I’ve Tried: Set up the reverse proxy with two endpoints for OpenDLP and the image classifier. Configured the image classifier to return a classification result (sensitive or non-sensitive). Tried to figure out how to integrate the classification results into OpenDLP’s file handling process, but I'm unsure how to communicate these results from the image classifier back into OpenDLP.

What I’m Looking For:
Guidance on how to use API calls or any other method to pass the classification results (sensitive or non-sensitive) from the image classifier to OpenDLP. Suggestions on best practices for integrating a Python-based ML classifier with a PHP-based LAMP stack (OpenDLP). Recommendations for managing the reverse proxy setup and ensuring seamless communication between OpenDLP and the image classifier.

Continue reading...