moshi-Kassim/model-for-the-early-detection-of-maize-leaf-diseases-in-tanzania
Domaine:
agricultureType de record:
modelsoftware
Créateur:
mos
Hôte:
# Maize Leaf Disease Detection Dashboard
Streamlit web dashboard for real-time diagnosis of maize leaf diseases using a trained Custom CNN model.
## Project
**Developing a Predictive Deep Learning Model for the Early Detection of Maize Leaf Diseases in Tanzania**
Detects four classes:
- Healthy
- Common Rust (*Puccinia sorghi*)
- Northern Leaf Blight (*Exserohilum turcicum*)
- Cercospora Leaf Spot (*Cercospora zeae-maydis*)
## Requirements
- Python 3.10+
- Trained model: `model/corn_disease_cnn.h5` (included)
## Setup
```bash
cd "python dashboard"
pip install -r requirements.txt
```
## Run the dashboard
There are **two separate apps**:
| Dashboard | File | Port | For |
|---|---|---|---|
| **User Dashboard** | `user_app.py` | 8501 | Farmers — upload, top disease, management only |
| **Admin Dashboard** | `app.py` | 8502 | Technical — detailed diagnosis, metrics, about |
**User Dashboard (farmers):**
```bash
streamlit run user_app.py
```
Or double-click `run_user.bat` / `run.bat`
**Admin Dashboard (technical):**
```bash
streamlit run app.py --server.port 8502
```
Or double-click `run_admin.bat`
- User app: `localhost`
- Admin app: `localhost`
Both apps **auto-load** the AI model on startup. Diagnosis runs **automatically** when you upload an image.
## Usage
**User Dashboard**
1. Upload a maize leaf image (JPG/PNG)
2. View the detected disease and recommended management
**Admin Dashboard**
1. Open **Diagnose** and upload a real maize leaf image for full probabilities and symptoms
2. Open **Performance** for model evaluation metrics
3. Open **About** for project information
> For panel presentation, use **live uploads only** (no pre-loaded sample images). Sample files in `assets/sample_images/` are kept for local developer testing only.
## Project structure
```
python dashboard/
├── user_app.py # User dashboard (farmers)
├── app.py # Admin dashboard (technical)
├── run_user.bat # Start …
Similaires
A deep learning model for early detection of maize diseases in TanzaniaDevelopment of Deep Learning Model for early detection of maize diseases in TanzaniaDeep learning models for the early detection of maize streak virus and maize lethal necrosis diseases in TanzaniaAn Interpretable Deep Learning for Early Detection and Diagnosis of Wheat Leaf DiseasesHyperspectral Imaging-Based Deep Learning Framework for Early Maize Leaf Disease DetectionA Mobile-Integrated Deep Learning Framework for Early Detection of Maize Diseases