Accurate estimation of crop yield predictions is of great importance for food security under the impact of climate change. We propose a data-driven crop model that combines the knowledge advantage of process-based modeling and the computational advantage of data-driven modeling. The proposed model tracks the daily biomass accumulation process during the maize growing season and uses daily produced biomass to estimate the final grain yield. Computational studies using crop yield, field location, genotype and corresponding environmental data were conducted in the US Corn Belt region from 1981 to 2020. The results suggest that the proposed model can achieve an accurate prediction performance with a 7.16% relative root-mean-square-error of average yield in 2020 and provide scientifically explainable results. The model also demonstrates its ability to detect and separate interactions between genotypic parameters and environmental variables. Additionally, this study demonstrates the potential value of the proposed model in helping farmers achieve higher yields by optimizing seed selection.

Plant breeders make selection decisions based on multiple traits, such as yield, plant height, flowering time, and disease resistance. A commonly used approach in multi-trait genomic selection is index selection, which assigns weights to different traits relative to their economic importance. However, classical index selection only optimizes genetic gain in the next generation, requires some experimentation to find weights that lead to desired outcomes, and has difficulty optimizing nonlinear breeding objectives. Multi-objective optimization has also been used to identify the Pareto frontier of selection decisions, which represents different trade-offs across multiple traits. We propose a new approach, which maximizes certain traits while keeping others within desirable ranges. Optimal selection decisions are made using a new version of the look-ahead selection (LAS) algorithm, which was recently proposed for single-trait genomic selection, and achieved superior performance with respect to other state-of-the-art selection methods. To demonstrate the effectiveness of the new method, a case study is developed using a realistic data set where our method is compared with conventional index selection. Results suggest that the multi-trait LAS is more effective at balancing multiple traits compared with index selection.

New genotyping technologies have made large amounts of genotypic data available for plant breeders to use in their efforts to accelerate the rate of genetic gain. Genomic selection (GS) techniques allow breeders to use genotypic data to identify and select, for example, plants predicted to exhibit drought tolerance, thereby saving expensive and limited field-testing resources relative to phenotyping all plants within a population. A major limitation of existing GS approaches is the trade-off between short-term genetic gain and long-term potential. Some approaches focus on achieving short-term genetic gain at the cost of reduced genetic diversity necessary for long-term gains. In contrast, others compromise short-term progress to preserve long-term potential without consideration of the time and resources required to achieve it. Our contribution is to define a new “look-ahead” metric for assessing selection decisions, which evaluates the probability of achieving high genetic gains by a specific time with limited resources. Moreover, we propose a heuristic algorithm to identify optimal selection decisions that maximize the look-ahead metric. Simulation results demonstrate that look-ahead selection outperforms other published selection methods.

Precise in-season corn grain yield estimates enable farmers to make real-time accurate harvest and grain marketing decisions minimizing possible losses of profitability. A well developed corn ear can have up to 800 kernels, but manually counting the kernels on an ear of corn is labor-intensive, time consuming and prone to human error. From an algorithmic perspective, the detection of the kernels from a single corn ear image is challenging due to the large number of kernels at different angles and very small distance among the kernels. In this paper, we propose a kernel detection and counting method based on a sliding window approach. The proposed method detects and counts all corn kernels in a single corn ear image taken in uncontrolled lighting conditions. The sliding window approach uses a convolutional neural network (CNN) for kernel detection. Then, a non-maximum suppression (NMS) is applied to remove overlapping detections. Finally, windows that are classified as kernel are passed to another CNN regression model for finding the $(x,y)$ coordinates of the center of kernel image patches. Our experiments indicate that the proposed method can successfully detect the corn kernels with a low detection error and is also able to detect kernels on a batch of corn ears positioned at different angles.

Goiffon et al. introduce the optimal population value (OPV) as a new metric for genomic selection. OPV is based on the maximum possible haploid value...Genomic selection (GS) identifies individuals for inclusion in breeding programs based on the sum of their estimated marker effects or genomic estimated breeding values (GEBVs). Due to significant correlation between GEBVs and true breeding values, this has resulted in enhanced rates of genetic gain as compared to traditional methods of selection. Three extensions to GS, weighted genomic selection (WGS), optimal haploid value (OHV) selection, and genotype building (GB) selection have been proposed to improve long-term response, and to facilitate the efficient development of doubled haploids. In separate simulation studies, these methods were shown to outperform GS under various assumptions. However, further potential for improvement exists. In this paper, optimal population value (OPV) selection is introduced as selection based on the maximum possible haploid value in a subset of the population. Instead of evaluating the breeding merit of individuals as in GS, WGS, and OHV selection, the proposed method evaluates the breeding merit of a set of individuals as in GB. After testing these selection methods extensively, OPV and GB selection were found to achieve greater responses than GS, WGS, and OHV, with OPV outperforming GB across most percentiles. These results suggest a new paradigm for selection methods in which an individual’s value is dependent upon its complementarity with others.