Farmer-oriented Flask and PyTorch application for crop disease detection, local guidance, and report generation. Built for practical use with image upload, prediction flow, Marathi-ready UX, weather context, and offline-friendly startup path.
- Accepts plant leaf images and runs disease prediction through local model inference.
- Returns disease label, confidence, and guidance for next action.
- Adds local weather context and market-oriented support data.
- Generates downloadable reports for field use and follow-up.
- Supports simple Windows startup through
START.bat.
- Python
- Flask
- PyTorch
- Pillow
- NumPy
Plant-Disease-Recognition-System/
app.py
backend/
models/
static/
templates/
uploadimages/
START.bat
requirements.txt
Requirements:
- Python 3
- Model file at
models/plant_disease_model_1_latest.pt
Install dependencies:
pip install -r requirements.txtStart app:
START.batManual fallback:
python app.pyDefault local URL:
http://127.0.0.1:5000
- Turns raw model output into usable guidance for farmers.
- Keeps workflow simple for local laptop deployment.
- Extends disease prediction with weather and report context instead of one-screen classification only.
- Main entrypoint:
app.py - Startup helper:
START.bat - Inference and service logic:
backend/ - Model asset expected locally; repo does not auto-download it