Grasision: Actionable Intelligence Tool for Sustainable Grassland Production

Grassision: Actionable Intelligence Tool for Sustainable Grassland Production Main goal Improve the analysis accuracy and precision of the algorithm for image classification and segmentation by refining it using new training data and ground-truth samples taken on various grass fields. In parallel, secure applicable IP rights and industrial partnerships. Secondary objectives 1. Develop the software architecture and user interface to provide the basis for a stable and user-friendly software for pilot-testing. 2. Process the available color image database and acquire new images for validation. 3. Sets the foundation for a subsequent funding application for a commercialization project. Aerial imagery will be processed through artificial intelligence techniques, including unsupervised and supervised methods to estimate plant features on grassland canopies, such as grass coverage, bare soil coverage, dead-material (e.g. straw, stones), and leaf area index. Available tools, effective on row- and horticultural crops, will be tested via transfer knowledge techniques. This tool will provide farmers with critical input to take cost-effective and responsible measures for fertilization, sward renovation and harvesting. In addition to the research results, we will secure applicable IP rights and better understand market dynamics for the commercial potential of our technology, while supporting NIBIO's ongoing research on precision agriculture and forage production.