Weedy Rice DSS
An applied research platform for agricultural segmentation and decision support.
Compute Split: Tier A Kaggle vs Tier B AWS
Weedy Rice DSS is a completed cross-season domain-gap study in UAV weedy-rice segmentation, built around a defended thesis. It measures the gap between wet- and dry-season imagery, then tests whether three unsupervised domain-adaptation methods — FDA, DAFormer and HRDA — close it, across 24 pre-registered training runs on a two-tier compute split. They do not: every method performs worse than no adaptation at all, and the study reports that negative result rather than burying it. A dormant product layer (Python backend, React/Vite UI, annotation tooling) was built earlier and is retained but is not part of the research path.
Overview
The Weedy Rice Decision Support System (DSS) is an applied research platform focused on AI-driven agricultural image segmentation. It bridges the gap between heavy machine learning workflows and an accessible interface.
Architecture
A dual-layer stack combining a Python 3.11 backend for AI processing and model training with a React (Vite) frontend. It utilizes specific Docker bind mounts for accessing massive external imagery datasets locally, while maintaining an AWS Lambda Hybrid deployment pattern for its production infrastructure. It keeps UI state and model logic cleanly separated.
Problem It Solves
Processing, segmenting, and analyzing specialized agricultural imagery requires dedicated, heavy infrastructure that isn't suited for standard web APIs. This system provides the dedicated environment to train models and visualize the results.
Current State
MVP operational. The separation between the backend AI stack, the React UI, and the segmentation-training workflows is fully aligned and documented as a cohesive stack.