7-Segment Display Digit Recognition

Introduction

This project is a prediction app that utilizes AI computer vision to recognize digits on a 7-segment display. The app is built using a FastAPI backend and a Nuxt 3 frontend. The backend receives images of 7-segment displays, processes them using a trained AI model, and sends the predictions to the frontend. The frontend displays the prediction and maintains a history of all previous predictions.

Project Structure

The project is divided into two main folders:

• ai: Contains the AI model and FastAPI backend.
• client: Contains the Nuxt 3 frontend.

AI Model

The AI model is trained on a dataset of images of 7-segment displays and is capable of recognizing digits (0-9). The model is implemented using the Ultralytics YOLOv8 library.

FastAPI Backend

The FastAPI backend:

• Receives images of 7-segment displays.
• Processes them using the AI model.
• Sends the predictions to the frontend.
• Is implemented using the FastAPI framework and is deployed on a local server.

Nuxt 3 Frontend

The Nuxt 3 frontend:

• Displays the prediction.
• Maintains a history of all previous predictions.
• Is implemented using the Nuxt 3 framework and is deployed on a local server.

Features

• Image Upload: Users can upload images of 7-segment displays to the app.
• Prediction: The app displays the predicted digit on the 7-segment display.
• Prediction History: The app maintains a history of all previous predictions.

Requirements

• Python 3.9+
• FastAPI
• Ultralytics YOLOv8
• Nuxt 3
• Node.js 14+

Installation

1. Clone the repository to your local machine:

   git clone <repository-url>
   cd <repository-folder>

2. Install the required dependencies:

   cd client && npm i

3. Start the FastAPI backend:

   cd ai && fastapi dev main.py

4. Start the Nuxt 3 frontend:

   cd client && npm run dev

Usage

1. Open the app in your web browser at http://localhost:3000
2. Upload an image of a 7-segment display.
3. The app will display the predicted digit.
4. The app will maintain a history of all previous predictions.

API Documentation

The FastAPI backend provides a RESTful API for interacting with the app. The API is documented using Swagger and can be accessed at:

• http://localhost:8000/docs

Commit Messages

Commit messages follow the conventional commit message format.

Branching Strategy

The project uses a feature-branch workflow. New features are developed in separate branches and merged into the main branch when complete.

To check more over my workflow and the train of thoughts check out the diagrams i have drawn out while planning this small project: Figjam Board

API Endpoints

The FastAPI backend provides the following API endpoints:

• POST /predict: Receives an image of a 7-segment display and returns the prediction consisting of x and y coordinates, class, and confidence level for each digit.

Model Training

The AI model is trained on a dataset of images of 7-segment displays using the Ultralytics YOLOv8 library.

Model Evaluation

The AI model is evaluated using metrics such as mAP50, mAP50-95.

Model Deployment

The AI model is deployed on a local server using the FastAPI framework.

Future Work

• Improve the accuracy of the AI model.
• Add support for recognizing multiple digits on a single 7-segment display.