Learn the steps to build an app that detects crop diseases

Laurence Moroney, a TensorFlow Developer Advocate at Google, outlines how developers can create an Android app capable of detecting crop diseases using various Google Developer tools, showcased at DevFest 2020. The process begins with building an Android app utilizing Kotlin, CameraX, and MLKit for on-device machine learning to analyze plant images. Gus Martins further enhances the app's capabilities by setting up a machine learning model with TensorFlow and TensorFlow Lite for disease detection in bean plants. Annyce Davis updates the app to use TensorFlow Lite for real-time inference on plant health. To transform this demo into a fully functional app, Todd Kerpelman suggests integrating Firebase for analytics and A/B testing, while Di Dang emphasizes the importance of responsible AI design. Paul Kinlan adds the web dimension by creating a Progressive Web App utilizing TensorFlow.js for browser-based machine learning. Lastly, Puuja Rajan highlights the benefits of open-sourcing the project to foster community collaboration and enhancement. This initiative demonstrates the collaborative use of Google's tools in building a minimal viable product with a roadmap for completion.