Journal ArticleSoybean maturity prediction using two‐dimensional contour plots from drone‐based time series imageryKim, Bitgoeul [Department of Computer Engineering Iowa State University Ames Iowa USA]; Blair, Samuel W [Department of Agronomy Iowa State University Ames Iowa USA]; Jubery, Talukder Z [Department of Mechanical Engineering Iowa State University Ames Iowa USA]; Sarkar, Soumik [Department of Mechanical Engineering Iowa State University Ames Iowa USA]; Singh, Arti [Department of Agronomy Iowa State University Ames Iowa USA] (ORCID:0000000161919238); Singh, Asheesh K [Department of Agronomy Iowa State University Ames Iowa USA] (ORCID:000000027522037X); Ganapathysubramanian, Baskar [Department of Mechanical Engineering Iowa State University Ames Iowa USA] (ORCID:0000000289314852)<title>Abstract</title> <p>Plant breeding programs require assessment and understanding of days to maturity for accurate selection and placement of entries in appropriate tests. Soybean [<italic>Glycine max</italic>(L.) Merr.] breeding programs, in the early stages of the breeding pipeline, assign relative maturity ratings to experimental varieties that indicate their suitable maturity zones. Traditionally, the estimation of maturity rating value has involved breeders manually inspecting fields and assessing maturity value visually. This approach relies heavily on expert judgment, making it subjective and demanding considerable time and effort. This study aimed to develop a machine learning (ML) model for evaluating soybean maturity using uncrewed aerial system (UAS)–based time series imagery. Images were captured at 3‐day intervals, beginning as the earliest varieties started maturing and continuing until the last varieties fully matured. The data collected for this experiment consisted of 22,043 plots collected across 3 years and represent relative maturity groups 1.6–3.9. We utilized contour plot images extracted from the time series UAS imagery as input for a neural network model. This contour plot approach encoded the temporal and spatial variation within each plot into a single image. A deep learning model was trained to utilize this contour plot to predict maturity ratings. This model demonstrates a significant improvement in accuracy and robustness, achieving up to 85% accuracy. The predictive model offers a scalable, objective, and efficient means of assessing crop maturity, enabling phenomics and ML approaches to reduce the reliance on manual inspection and subjective assessment, thereby saving time and resources in a breeding program.</p>The Plant Phenome Journal2025-12-0110673221The Plant Phenome Journal812578-2703https://doi.org/10.1002/ppj2.700361954556National Science Foundation