Maize Yield Prediction at an Early Developmental Stage Using Multispectral Images and Genotype Data for Preliminary Hybrid Selection

Abstract

Assessing crop production in the field often requires breeders to wait until the end of the season to collect yield-related measurements, limiting the pace of the breeding cycle. Early prediction of crop performance can reduce this constraint by allowing breeders more time to focus on the highest-performing varieties. Here, we present a multimodal deep learning model for predicting the performance of maize (Zea mays) at an early developmental stage, offering the potential to accelerate crop breeding. We employed multispectral images and eight vegetation indices, collected by an uncrewed aerial vehicle approximately 60 days after sowing, over three consecutive growing cycles (2017, 2018 and 2019). The multimodal deep learning approach was used to integrate field management and genotype information with the multispectral data, providing context to the conditions that the plants experienced during the trial. Model performance was assessed using holdout data, in which the model accurately predicted the yield (RMSE 1.07 t/ha, a relative RMSE of 7.60% of 16 t/ha, and R2 score 0.73) and identified the majority of high-yielding varieties, outperforming previously published models for early yield prediction. The inclusion of vegetation indices was important for model performance, with a normalized difference vegetation index and green with normalized difference vegetation index contributing the most to model performance. The model provides a decision support tool, identifying promising lines early in the field trial.

Estimation of Maize Yield and Flowering Time Using Multi-Temporal UAV-Based Hyperspectral Data

“…A partial least squares regression (PLSR) model trained with the time-series BNDVI achieved an average accuracy of r = 0.51 for grain yield and r = 0.69-0.70 for the flowering time [36]. Danilevicz et al (2021) used multispectral VIs, genotype information, and field management data to predict maize yield of G2F breeding trails, hitting an accuracy of R 2 = 0.73 [44]. Herrmann et al (2020) assessed maize yield over 19 maize hybrids using VIs extracted from a super-spectral sensor mounted on a UAV, reaching an R 2 of 0.73 for grain yield prediction [45].…”

Application of crop wild relatives in modern breeding: An overview of resources, experimental and computational methodologies

“…Similarly, Ma et al., 2018 ., successfully developed a ML model to predict eight phenotypic traits among 2000 wheat individuals using 33,709 DArT (Diversity Array Technology) markers ( Ma et al., 2018 ). ML is now also being used to predict mature yield in early development using a combination of image and genotype data ( Danilevicz et al., 2021 ; Danilevicz et al., 2022 ). Recently ML models were developed for identification of core and dispensable genes in Oryza sativa L. and Brachypodium distachyon (L.) P. Beauv.…”

Artificial intelligence framework for modeling and predicting crop yield to enhance food security in Saudi Arabia

“…According to the suggested multimodal deep learning framework, the intermediate-level feature fusion DNN framework outperformed its input-level feature fusion DNN framework in terms of prediction accuracy, spatial adaptability, and resilience. Multimodal deep learning was employed by Danilevicz et al (2021) by integrating tab-DNN, sp-DNN, two linear layers of fusion, and ReLU. The weights from the last layers of tab-DNN and sp-DNN were combined for the fusion module input.…”