TED-QSRT 0.1.4

A modular QSR Order Verification Python Package

TED_QSRT

A modular, highly accurate computer vision verification system designed for Quick Service Restaurants (QSRs). This package compares a camera image of a packed food tray with OCR text extracted from an order receipt to ensure the order is correct.

Installation

You can install this directly from source:

cd fastfood_verifier_project
pip install -e .

Or build a distribution wheel for deployment on other servers:

# Build the wheel
python setup.py sdist bdist_wheel

# Install the wheel (replace with actual filename in dist/)
pip install dist/TED_QSRT-0.1.0-py3-none-any.whl

Command Line Usage

Once installed, you can trigger the AI verification pipeline directly from the terminal. It accepts two image paths and returns a JSON verification result.

ted-qsrt --item inputs/camera_view.jpg --receipt inputs/order_receipt.jpg

Alternatively, you can run the module directly:

python -m TED_QSRT --item inputs/camera_view.jpg --receipt inputs/order_receipt.jpg

REST API Server Usage

This package also comes with a built-in FastAPI server to immediately work with web interfaces or edge/mobile devices (iOS/Android).

Start the API Server:

uvicorn TED_QSRT.api.server:app --host 0.0.0.0 --port 8000

Endpoints

POST /verify Accepts two multipart image uploads: item_image and receipt_image. Returns a JSON object containing the comparison results.

Example Request:

curl -X POST "http://localhost:8000/verify" \
  -H "accept: application/json" \
  -H "Content-Type: multipart/form-data" \
  -F "item_image=@inputs/camera_view.jpg" \
  -F "receipt_image=@inputs/order_receipt.jpg"