🌾 Smart Crop Disease Detection System

An AI-powered web application built with Django that detects crop diseases from leaf images using deep learning. Designed to assist farmers and agricultural professionals with quick and accurate disease identification and recommendations.

---

🚀 Features

• 🌿 Upload leaf images and detect plant diseases with high accuracy.
• 🤖 Integrated trained ML model (TensorFlow/TensorFlow Lite).
• 🗂️ Stores prediction records in PostgreSQL with advanced search & filtering (Elasticsearch-ready).
• ☁️ AWS S3 integration for scalable image storage.
• 🔐 User authentication and role-based access control (OAuth support included).
• ⚙️ Background ML inference with Celery (future enhancement).
• 📊 Admin dashboard for managing data.(Currently working on it)
• 📦 Dockerized for easy deployment.
• 🧪 Unit-tested with high coverage.

---

🛠️ Tech Stack

Category	Tools / Technologies
Backend	Django, Django REST Framework
Machine Learning	TensorFlow / TensorFlow Lite, OpenCV
Database	PostgreSQL
Cloud Storage	AWS S3
DevOps	Docker, Git, GitHub Actions (CI/CD)
Authentication	Django Allauth / OAuth2 (Google/Facebook)
Future Upgrades	Celery + Redis (async tasks), Elasticsearch

---

🖼️ Demo

(SOON WILL POST HERE)

Try it live: https://smartcrop.example.com (if deployed)

---

📂 Project Structure

smart_crop/ │ ├── detection/ # App for leaf image handling & prediction │ ├── models.py # LeafImage model │ ├── views.py # Upload & prediction logic │ └── ml/ # Model files (TFLite or H5) │ ├── templates/ # HTML templates (Bootstrap + custom) ├── static/ # Static files (CSS/JS) ├── media/ # Uploaded images (S3 synced) ├── core/ # Project settings and URLs │ ├── settings.py │ ├── settings_local.py # Secrets (gitignored) │ ├── Dockerfile ├── docker-compose.yml ├── requirements.txt └── README.md

---

🧠 How It Works

1. Upload a crop/leaf image via the web interface.
2. The backend sends the image to a pre-trained ML model.
3. The model classifies the disease and returns a prediction.
4. The result is stored in PostgreSQL and optionally searchable via Elasticsearch.

---

📸 Sample Predictions

Image	Prediction	Confidence
tomato_leaf_01.jpg	Tomato Early Blight	92.3%
potato_leaf_02.jpg	Potato Late Blight	89.7%

---

⚙️ Setup Instructions

Prerequisites

• Python 3.10+
• PostgreSQL
• AWS S3 bucket (configured)
• Docker (optional but recommended)

1. Clone the Repository

git clone https://github.com/yourusername/smart_crop.git
cd smart_crop

2. Install Dependencies

pip install -r requirements.txt

3. Configure Environment

Create settings_local.py for sensitive data (DB, AWS, Secret Keys):

# settings_local.py
DATABASES = {
    'default': {
        'ENGINE': 'django.db.backends.postgresql',
        'NAME': 'smart_crop_db',
        'USER': 'postgres',
        'PASSWORD': 'yourpassword',
        'HOST': 'localhost',
        'PORT': '5432',
    }
}

AWS_ACCESS_KEY_ID = 'your-access-key'
AWS_SECRET_ACCESS_KEY = 'your-secret-key'
AWS_STORAGE_BUCKET_NAME = 'your-bucket-name'

4. Run Migrations

python manage.py makemigrations
python manage.py migrate

5. Create Superuser

python manage.py createsuperuser

6. Run the Server

python manage.py test

📦 Docker Support

docker-compose up --build

✨ Future Enhancements ✅ Integrate Celery for background model inference.

✅ Add Elasticsearch for full-text search on prediction records.

✅ Advanced filtering (by crop type, confidence, date).

✅ Model retraining from admin panel.

✅ Multilingual support (Hindi, etc.).

🤝 Contributing Contributions are welcome! Please fork the repo and submit a pull request.

📜 License MIT License © HARSH GANGWAR

📧 Contact Author: HARSH GANGWAR

Email: hgangwar07@gmail.com

LinkedIn: www.linkedin.com/in/harshgangwar07