Search for: All records

Abstract Insect pests significantly impact global agricultural productivity and crop quality. Effective integrated pest management strategies require the identification of insects, including beneficial and harmful insects. Automated identification of insects under real-world conditions presents several challenges, including the need to handle intraspecies dissimilarity and interspecies similarity, life-cycle stages, camouflage, diverse imaging conditions, and variability in insect orientation. An end-to-end approach for training deep-learning models, InsectNet, is proposed to address these challenges. Our approach has the following key features: (i) uses a large dataset of insect images collected through citizen science along with label-free self-supervised learning to train a global model, (ii) fine-tuning this global model using smaller, expert-verified regional datasets to create a local insect identification model, (iii) which provides high prediction accuracy even for species with small sample sizes, (iv) is designed to enhance model trustworthiness, and (v) democratizes access through streamlined machine learning operations. This global-to-local model strategy offers a more scalable and economically viable solution for implementing advanced insect identification systems across diverse agricultural ecosystems. We report accurate identification (>96% accuracy) of numerous agriculturally and ecologically relevant insect species, including pollinators, parasitoids, predators, and harmful insects. InsectNet provides fine-grained insect species identification, works effectively in challenging backgrounds, and avoids making predictions when uncertain, increasing its utility and trustworthiness. The model and associated workflows are available through a web-based portal accessible through a computer or mobile device. We envision InsectNet to complement existing approaches, and be part of a growing suite of AI technologies for addressing agricultural challenges. 

Abstract Developments in genomics and phenomics have provided valuable tools for use in cultivar development. Genomic prediction (GP) has been used in commercial soybean [Glycine maxL. (Merr.)] breeding programs to predict grain yield and seed composition traits. Phenomic prediction (PP) is a rapidly developing field that holds the potential to be used for the selection of genotypes early in the growing season. The objectives of this study were to compare the performance of GP and PP for predicting soybean seed yield, protein, and oil. We additionally conducted genome‐wide association studies (GWAS) to identify significant single‐nucleotide polymorphisms (SNPs) associated with the traits of interest. The GWAS panel of 292 diverse accessions was grown in six environments in replicated trials. Spectral data were collected at two time points during the growing season. A genomic best linear unbiased prediction (GBLUP) model was trained on 269 accessions, while three separate machine learning (ML) models were trained on vegetation indices (VIs) and canopy traits. We observed that PP had a higher correlation coefficient than GP for seed yield, while GP had higher correlation coefficients for seed protein and oil contents. VIs with high feature importance were used as covariates in a new GBLUP model, and a new random forest model was trained with the inclusion of selected SNPs. These models did not outperform the original GP and PP models. These results show the capability of using ML for in‐season predictions for specific traits in soybean breeding and provide insights on PP and GP inclusions in breeding programs. 

Abstract Soybean (Glycine max[L.] Merr.) production is susceptible to biotic and abiotic stresses, exacerbated by extreme weather events. Water limiting stress, that is, drought, emerges as a significant risk for soybean production, underscoring the need for advancements in stress monitoring for crop breeding and production. This project combined multi‐modal information to identify the most effective and efficient automated methods to study drought response. We investigated a set of diverse soybean accessions using multiple sensors in a time series high‐throughput phenotyping manner to: (1) develop a pipeline for rapid classification of soybean drought stress symptoms, and (2) investigate methods for early detection of drought stress. We utilized high‐throughput time‐series phenotyping using unmanned aerial vehicles and sensors in conjunction with machine learning analytics, which offered a swift and efficient means of phenotyping. The visible bands were most effective in classifying the severity of canopy wilting stress after symptom emergence. Non‐visual bands in the near‐infrared region and short‐wave infrared region contribute to the differentiation of susceptible and tolerant soybean accessions prior to visual symptom development. We report pre‐visual detection of soybean wilting using a combination of different vegetation indices and spectral bands, especially in the red‐edge. These results can contribute to early stress detection methodologies and rapid classification of drought responses for breeding and production applications. 

Abstract Insect pests cause significant damage to food production, so early detection and efficient mitigation strategies are crucial. There is a continual shift toward machine learning (ML)‐based approaches for automating agricultural pest detection. Although supervised learning has achieved remarkable progress in this regard, it is impeded by the need for significant expert involvement in labeling the data used for model training. This makes real‐world applications tedious and oftentimes infeasible. Recently, self‐supervised learning (SSL) approaches have provided a viable alternative to training ML models with minimal annotations. Here, we present an SSL approach to classify 22 insect pests. The framework was assessed on raw and segmented field‐captured images using three different SSL methods, Nearest Neighbor Contrastive Learning of Visual Representations (NNCLR), Bootstrap Your Own Latent, and Barlow Twins. SSL pre‐training was done on ResNet‐18 and ResNet‐50 models using all three SSL methods on the original RGB images and foreground segmented images. The performance of SSL pre‐training methods was evaluated using linear probing of SSL representations and end‐to‐end fine‐tuning approaches. The SSL‐pre‐trained convolutional neural network models were able to perform annotation‐efficient classification. NNCLR was the best performing SSL method for both linear and full model fine‐tuning. With just 5% annotated images, transfer learning with ImageNet initialization obtained 74% accuracy, whereas NNCLR achieved an improved classification accuracy of 79% for end‐to‐end fine‐tuning. Models created using SSL pre‐training consistently performed better, especially under very low annotation, and were robust to object class imbalances. These approaches help overcome annotation bottlenecks and are resource efficient.