Huawei Tech4City Challenge - Team Predict Guardian Submission

[hanlim83]

Fire/Smoke Detection by team Predict Guardian

Our Code Submission is a full-stack AI web application that detects the presence of flames or smoke in the user's uploaded images.

How to run our application

Docker (Recomended)

1. Run the docker container via the following command docker run -d -p 8000:8000 --name predict-guardian hanlim82/huawei-tech4city:latest

Locally

1. cd into the backend directory
2. Create a python virtual environment (venv) python -m venv venv
3. Activate the newly created virtual environment
4. Install the required dependencies via pip pip install -r .\requirements.txt
5. Run the backend by running uvicorn app:app

NOTE: If you are running Debian/Ubuntu, before Step 4, please run this command apt-get update && apt-get install ffmpeg libsm6 libxext6 -y

How can I access the UI/API?

You will be able to access the user interface and the API via localhost:8000

The API Documentation can be accessed via localhost:8000/docs

Technologies Used

• FastAPI
• Ultralytics YOLOv8 - Customised Model For Fire and Smoke Detection
• Daisy UI
• Tailwind CSS
• Font Awesome

Team Members

We are all year 3 students from SMU School of Computing and Information Systems

• Juay Kai Xun (Team Leader)
• Hansen Lim (AI Specialist)
• Ashley Tan (Applications Lead)
• Syahmi Abbas (Solutions Lead)

[codeautopilot]

PR summary

This Pull Request introduces a full-stack AI web application developed by the Predict Guardian team for the Huawei Tech4City Challenge. The application detects the presence of flames or smoke in user-uploaded images using a customized YOLOv8 model. The backend is built with FastAPI, and the frontend uses Daisy UI and Tailwind CSS. The application can be run using Docker or locally, and it includes endpoints for uploading images and retrieving past results.

Suggestion

1. Error Handling: Enhance error handling in the backend to provide more specific error messages.
2. Security: Implement security measures such as input validation and rate limiting to prevent abuse.
3. Scalability: Consider using a more scalable database solution if the application is expected to handle a large number of uploads.
4. Testing: Add unit and integration tests to ensure the reliability of the application.
5. Documentation: Improve documentation to include more detailed setup instructions and API usage examples.

Disclaimer: This comment was entirely generated using AI. Be aware that the information provided may be incorrect.

Current plan usage: 41.45%

Have feedback or need help?
Discord
Documentation
support@codeautopilot.com