This API provides a simple way to perform object detection using the YOLO (You Only Look Once) model implemented with Darknet, integrated with a Flask server. It allows users to send images to the server and receive JSON responses containing the result. Currently, this model can detect 80 objects
- Clone the repository
git clone https://github.com/aravind-tronix/object-detection-web-client.git
cd object-detection-web-client
- Install required packages by
pip install -r requirements.txt
- Download the model manually since it exceeds 100mb.
- Place the model file into the project root folder.
- Start the Flask server
python app.py
- Start the Front-end app >>> https://github.com/aravind-tronix/object-detection-webapp
The architecture of this API involves four main components:
- YOLO (You Only Look Once) is a state-of-the-art, real-time object detection system.
- Darknet is an open-source neural network framework written in C and CUDA.
- Project uses the YOLO model implemented with Darknet for efficient object detection.
- Flask is a lightweight WSGI web application framework in Python.
- Have integrated the YOLO model with a Flask server to create a RESTful API.
- The Flask server listens for incoming HTTPS requests, processes them, and returns JSON responses.
- Apache HTTP Server is a widely-used open-source web server software.
- Have deployed the Flask application behind Apache using mod_wsgi module to serve the API.
- Apache acts as a gateway, routing incoming requests to the Flask application.
- Hetzner Cloud is a cloud hosting provider offering scalable virtual private servers (VPS) and cloud infrastructure.
- The deployment of the Apache HTTP Server along with the Flask application is done on Hetzner Cloud infrastructure for reliable and scalable hosting.
View and run the API on postman:
The app is live at https://dev.aravind.one/