High Throughput Phenotyping (HTP) for Single Plant Using Unmanned Aerial System (UAS) Data in Cotton

Unmanned Aerial System (UAS) and sensor technology have made it possible to collect very high spatial and temporal resolution data. Advanced UAS data provide a great opportunity to identify individual plants and to extract plant-level attributes for high-throughput phenotyping (HTP) and precision agriculture. We collected RGB and multi-spectral data from UAS with high overlap (> 80%) at low altitude (<30m). The orthomosaic and Digital Surface Model (DSM) were generated by using Structure from Motion algorithm with Ground Control Point (GCP) for precision geo-referencing. We have optimized the use of UAS imagery data to identify individual plants desirable agronomic properties relative to plant architecture, plant growth, NDVI, and yield based on boll counts at season end. The single plant polygon was generated from the UAS data collected after cotton emerging and the attributes for each plant was extracted. This study will compare the attributes of single plants from UAS with yield.