DeepNude-Detector

Description

This project is a tool for detecting nudity in images, allowing users to analyze both local images and images from URLs. It utilizes the NudeNet library to classify and detect explicit content. Additionally, it offers an option to automatically censor images identified as inappropriate.

Features

• Detects inappropriate content in images.
• Allows users to choose between analyzing local images or downloading images from a URL.
• Automatically blurs images detected as inappropriate.
• Generates random names for downloaded and censored images to prevent file overwriting.

How It Works

1. The user can choose to analyze either a local image or download an image from a URL.
2. The system detects if the image contains explicit content based on a configurable minimum score.
3. If requested, the image is censored and saved locally with a randomly generated name in the format censored_<random_name>.jpg.

Requirements

• Python 3.x
• Libraries:
  • NudeNet
  • OpenCV
  • Pillow (PIL)
  • Requests

Contribution

Feel free to open issues and submit pull requests for improvements or bug fixes. This project is under active development, and feedback is always appreciated.