Development of a Low-Cost Automated Greenhouse Imaging System with Machine Learning-Based Processing for Evaluating Genetic Performance of Drought Tolerance in a Bentgrass Hybrid Population

Authors: KIM, YONGHYUN - Orise Fellow; Barnaby, Jinyoung; Warnke, Scott

Submitted to: Computers and Electronics in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/25/2024
Publication Date: 6/30/2024
Citation: Kim, Y., Barnaby, J.Y., Warnke, S.E. 2024. Development of a Low-Cost Automated Greenhouse Imaging System with Machine Learning-Based Processing for Evaluating Genetic Performance of Drought Tolerance in a Bentgrass Hybrid Population. Computers and Electronics in Agriculture. https://doi.org/10.1016/j.compag.2024.108896.
DOI: https://doi.org/10.1016/j.compag.2024.108896

Interpretive Summary: Evaluating a large population of plants for stress-tolerance traits can be laborious and time consuming, so is often done after the plants are showing extreme visible signs of stress. However, stress progression rates can vary widely in a population, and plants with valuable traits may be missed if evaluation occurs too late. To address this issue in turfgrass evaluation, ARS Scientists developed a low-cost automated greenhouse-based red/green/blue (RGB) imaging system and image processing platform that uses machine learning methods to evaluate drought tolerance in a bentgrass hybrid population. This system not only saves time and labor, but also allows data about the plant’s stress response to be combined with genomic information to facilitate mapping of these traits. This will lead to more effective selection of drought-resilient turfgrass germplasm and facilitates development of improved turf varieties for consumers.

Technical Abstract: Precise phenotypic assessment of large mapping populations, comprised of a few thousand plants, is time-consuming and labor-intensive. Depending on the trait being phenotyped, results may be subjectively assessed and variable due to environmental effects. This is particularly the case when evaluating genetic responses to abiotic stress which may involve multiple measurements over time as well as a range of induced stress levels. A machine learning (ML)-based imaging analysis system offers an efficient and precise means to capture temporal progression of stress symptoms within a large genetic population that can be interpreted through computer vision algorithms. Extreme weather events due to climate change threaten sustainable crop production and quality, and developing crop varieties with resilient yield and quality traits through breeding is needed. One of the major issues for turfgrass management is excessive water use for irrigation. In the face of climate change, water availability is becoming increasingly limited and more costly, and water conservation in turfgrass culture has become extremely important, and thus development of drought tolerant turfgrass germplasm is critical. In this study, we developed an automatic image analysis system through ML methods that can automatically process several hundred thousand images in a few hours. This system can automatically (1) capture images, (2) segment an image containing numerous plants into sub-images with individual plants, (3) label each image with the corresponding sample information, (4) quantify stress symptoms, and (5) classify breeding lines based on performance. Such artificial intelligence (AI)-based high-throughput digital phenotyping platforms can significantly increase the effectiveness of germplasm evaluation, quantitative trait locus (QTL) mapping and candidate gene identification to develop potential molecular markers that will aid in faster development of improved germplasm.