greenhealth

Empowering farmers with AI-driven plant disease diagnosis, bilingual treatment guidance, and downloadable advisory reports to prevent crop loss in real time.

Comment

Millions of farmers globally lose 20% to 40% of their annual crop yields to preventable plant diseases and delayed diagnostics. In many regions, access to agricultural experts, agronomists, or specialized diagnostic labs is costly, slow, or unavailable.

We were inspired to build a universally accessible, AI-powered crop health platform that delivers instant disease detection, actionable treatment advisories, and print-ready field reports—empowering farmers worldwide to protect their crops and secure global food supplies.

AgriShield AI is an end-to-end crop disease diagnosis and advisory platform designed for global agricultural impact:

  • 📸 AI Disease Diagnosis: Detects diseases across key crops (Tomato, Potato, Rice, Wheat, Maize, Cotton, Chilli, Apple) using symptom analysis and visual indicators, delivering severity ratings and confidence scores in real time.
  • 🌐 Global Accessibility Engine: Built with a primary English interface and multi-language localization support (including regional options like Hindi) for seamless adoption across different farming communities.
  • 📘 Verified Actionable Advisories: Provides comprehensive guidelines on immediate treatment steps, management practices, and long-term prevention methods sourced from verified agricultural research institutes.
  • 📄 Downloadable PDF Field Reports: Generates professional, offline-ready PDF advisory reports that farmers can save, print, or share with local agricultural extension officers.
  • 🔌 Extensible REST APIs: Features full RESTful API endpoints (/api/crops/, /api/diseases/, /api/diagnosis/, /api/reports/) ready for mobile application and IoT sensor integrations.

- Backend & Logic: Built with Python and Django utilizing robust ORM data modeling for symptoms, crops, diseases, and verified knowledge sources.

  • AI/ML Engine: Custom diagnostic classifier mapping user-observed symptoms and leaf images against verified plant pathology field guidelines.
  • Dynamic PDF Generator: Powered by ReportLab to generate custom-styled PDF field advisory reports on the fly.
  • Frontend & Infrastructure: Responsive HTML5/CSS3/JavaScript interface with font-awesome indicator badges, WhiteNoise static asset pipeline, and multi-language engine architecture.
  • Deployment & Cloud: Serverless architecture optimized for Vercel with pre-seeded database initialization, Cloudinary media storage, and Docker containerization support.
  1. Serverless SQLite State Management: Vercel’s serverless functions operate on a read-only filesystem. We solved this by architecting a custom startup handler that initializes a pre-seeded SQLite database in /tmp/db.sqlite3 with proper write permissions on cold starts.
  2. Global Localization Architecture: Designing a clean database schema and dynamic translation pipeline that supports multi-language expansion without increasing API latency or duplicating queries.
  3. Lambda PDF Optimization: Optimizing ReportLab rendering memory overhead to ensure PDF advisories generate instantly inside serverless execution limits.

- Successfully deploying a full-stack Django application with pre-seeded agricultural knowledge to a zero-downtime serverless cloud platform.

  • Creating an intuitive, universally accessible tool that bridges the gap between complex plant pathology and practical farm management.
  • Delivering instant, sub-second disease diagnosis with clear actionable immediate and long-term remedies.
  • Best practices for optimizing Python and Django applications for serverless cloud environments (Vercel/AWS Lambda).
  • Building scalable multi-language internationalization (i18n) engines for global user bases.
  • Structuring scalable REST APIs for domain-specific knowledge bases and diagnostic workflows.
  • 📱 Global Mobile Application: Developing a React Native / Flutter mobile app with offline-first caching for farmers in remote regions with low connectivity.
  • 🛰️ IoT & Sensor Integration: Integrating soil moisture, weather data, and drone camera feeds for early automated disease outbreak warnings.
  • 🎙️ Voice-Assisted AI: Adding multi-lingual voice interaction (Speech-to-Text) so farmers can report crop symptoms hands-free.

Built With

Share this project:

Updates

posted an update —

We are excited to announce the initial release and cloud deployment of AgriShield AI (greenhealth)!

Key Highlights in this Build:

  • AI Crop Disease Diagnostics: Real-time symptom analysis and severity scoring across major global crops (Tomato, Potato, Rice, Wheat, Maize, Cotton, Chilli, Apple).
  • Global Accessibility Engine: Fully responsive web dashboard featuring multi-language localization support.
  • Dynamic PDF Field Advisories: Integrated ReportLab engine to generate downloadable, print-ready field reports.
  • Serverless Architecture: Production build deployed on Vercel with optimized REST APIs and SQLite caching. Feel free to check out our live demo and code repository. We welcome all feedback and suggestions!

Log in or sign up for Devpost to join the conversation.

Submission history