PPE Detection System👷‍♂️

This is a web-based Personal Protective Equipment (PPE) detection system designed to enhance workplace safety by identifying compliance with safety standards, such as wearing helmets, masks, and safety vests. The system leverages YOLOv8 (You Only Look Once) for real-time object detection and is deployed as a web application using React for the frontend and Flask for the backend.

Key Features

• PPE Detection: Detects and classifies PPE, including hardhats, masks, and safety vests, using YOLOv8.
• Industry-Specific Safety Information: Provides tailored safety messages based on the industry (e.g., Construction, Manufacturing) and missing PPE.
• Image Upload and Processing: Users can upload images, which are processed to identify the presence or absence of required PPE.
• Fast and Accurate: Leverages YOLOv8 for high-speed and precise detection.

Step-by-Step Working

1. Image Upload:

   • Users upload an image of workers through the web interface.
   • The user specifies the industry type to customize PPE requirements.

2. Image Processing and Detection:

   • The uploaded image is sent to the Flask backend.
   • YOLOv8, pre-trained on a dataset of PPE images, processes the image to detect and classify PPE items such as helmets, vests, and masks.

3. PPE Validation:

   • The system compares detected PPE against the mandatory requirements for the specified industry.
   • Missing PPE is identified.

4. Safety Message Generation:

   • Based on the missing PPE, the system generates safety warnings and recommendations tailored to the industry.

5. Results Display:

   • The results, including a list of detected PPE, missing PPE, and safety messages, are displayed on the web application.
   • The user can also view processed images with bounding boxes around detected PPE items.

Example Images

• Without PPE: An example showing a worker without the required PPE.
  

• With PPE: An example showing a worker with the required PPE.
  

Endpoints

1. POST /upload

Description:
Allows users to upload an image for PPE detection.

Parameters:

• file (file): The image file containing workers.
• industry (form data): The industry for which PPE requirements and safety warnings are calculated

Response:
Returns a JSON with:

• detected_ppe: List of PPE detected in the image.
• missing_ppe: List of PPE that is missing, based on the industry.
• safety_message: Safety warnings for each missing PPE.
• annotated_image: Image with bounding boxes around detected PPE items.