Small plot identification from video streams for high-throughput phenotyping of large breeding populations with unmanned aerial systems

摘要

In breeding and genetics programs, thousands of small plots in the size of a few square meters each having a uniquegenetic entry, are used to evaluate huge numbers of candidates and large mapping populations. Low-altitude remotesensing with unmanned aerial systems (UAS) can generate high geospatial resolution measurements of plants and enablehigh temporal resolution measurements through multiple crop growth stages. However, to identify individual plot fromaerial images robustly and automatically becomes a key challenge in high throughput phenotyping (HTP) using UAS. Inthis case study, we captured super high-resolution video clips of wheat canopies by a UAS at low altitude. We proposedan image processing pipeline for identifying individual plot from video frames. Preliminary results indicate the methodscan highly accelerate the process of linking genotypes to individual plot images and can be fully automated. Thisresearch provides a proof-of-concept and has broad implications of novel phenomics application of UAS that is scalableto tens-of-thousands of plots in crop breeding and genetic studies.