Detecting and Counting Panicles in Sorghum Images

Peder A. Olsen, Karthikeyan Natesan Ramamurthy, Javier Ribera, Yuhao Chen, Addie M. Thompson, Ronny Luss, Mitch Tuinstra, Naoki Abe

Abstract

Phenotyping, the process of measuring plant traits, plays a central role in plant breeding. However, traditional approaches are labor-intensive, time-consuming, costly, and error prone. Accurate, automated, high-throughput phenotyping can relieve a huge burden in the breeding pipeline. In this paper, we propose computer vision systems and approaches to annotate, detect, and count panicles (heads), a key phenotype, from aerial images of Sorghum crops. The annotation system allows the users to label panicles in Sorghum aerial images. This annotated data is used for learning by the panicle detection and counting algorithms. The proposed approaches were used with aerial imagery of 18 varieties of Sorghum crop collected at 6 different dates in the Midwestern United States. The detector has an AUC of over 0.98 and the counter has a mean absolute error of 2.66 without adapting to variety and 1.88 when using variety specific information. Our approaches are being adopted into a high-throughput phenotyping pipeline for accelerating Sorghum breeding.
BibTeX
@inproceedings{olsen2018detecting,
  author = {Peder A. Olsen and Karthikeyan Natesan Ramamurthy and Javier Ribera and Yuhao Chen and Addie M. Thompson and Ronny Luss and Mitch Tuinstra and Naoki Abe},
  title = {{Detecting and Counting Panicles in Sorghum Images}},
  booktitle = {DSAA},
  year = {2018},
  doi = {10.1109/dsaa.2018.00052},
}

Loading related content...

Figure