Image Detection API

This application leverages the Ultralytics YOLOv8 object detection model to accurately identify and locate objects within an image. It provides a user interface for uploading an image and obtaining an annotated version that highlights and classifies various objects.

Features

• Image upload and processing
• Object detection using YOLOv8 model
• Annotates images with detected object classes, their bounding boxes and confidence scores using OpenCV (Open Source Computer Vision Library)

Getting Started

Prerequisites

• Docker

Installation

Clone the repository to your local machine:

git clone https://github.com/feldeh/image-detection-api

Navigate to the project's root directory:

cd image-detection-api

Running the Application

To run the application, you need to spin up the server and app Docker containers:

docker-compose up

App Usage

1. Open your browser and navigate to localhost:3000.
2. Upload an image.
3. Wait a few seconds until the processed image appears on the screen with the detected objects annotated.

API Reference

You can interact with the API endpoints using the Swagger UI documentation. The Swagger UI is automatically generated and can be accessed from the /docs endpoint.

Health check

  GET /

Returns "Ready" if the server is running.

Upload image

  POST /upload

Form Field	Type	Description
file	file	Required. Image file to upload

Upload an image file for processing. The image file should be in the form-data format with the key 'file'.

Built With

• FastAPI - Web framework used to build the API.
• Ultralytics YOLOv8 - Object detection model.
• OpenCV - Used for image processing.

Roadmap

• Enhance the object detection API endpoint to return the processed image.
• Integrate a front-end user interface to interactively use the API and visualize the results.