Optimizing data collection requirements for machine learning models in wild blueberry automation through the application of DALL-E 2

• Integration of AI-generated imagery, using DALL-E 2, into the data collection process for machine learning models in precision agriculture, with a focus on wild blueberry cultivation. • Augmenting traditional datasets with AI-generated images, the study sought to enhance the detection capabilities of models targeting key variables in wild blueberry production, including the identification of ripe berries, hair fescue weeds, and red leaf disease. • The results demonstrated that the combination of AI-generated and real images consistently outperformed models trained on either dataset alone. • By effectively combining AI-generated imagery with real data, the study offers a promising method for advancing precision agriculture, making data collection more efficient and models more reliable across various agricultural applications. This research developed a workflow to assess the viability of AI-generated imagery in training machine learning models for detecting ripe wild blueberries ( Vaccinium angustifolium Ait.), hair fescue weeds ( Festuca filiformis Pourr.), and red leaf disease ( Exobasidium vaccinii ). Ground truth images were collected and augmented with AI-generated variations using DALL-E 2 to expand the dataset. Models were trained on three datasets: ground truth, generated, and a combination (40% generated images). Evaluation metrics included precision, recall, mAP 50 , and mAP 50–95 , analyzed using ANOVA multiple mean comparisons and Tukey's HSD test (α = 0.05). For ripe wild blueberries, combination models achieved the highest performance across all metrics (mAP 50 : 0.834), significantly outperforming the ground truth model (mAP 50 : 0.806) in terms of mAP 50–95 (0.478 compared to 0.424). For hair fescue weeds, the combination dataset outperformed others with the highest mAP 50 (0.983), closely followed by the ground truth dataset (mAP 50 : 0.969). In detecting red leaf disease, the combination dataset showed the best performance (mAP 50 : 0.848 ± 0.140, mAP 50–95 : 0.607 ± 0.219), compared to the ground truth (mAP 50 : 0.615 ± 0.092, mAP 50–95 : 0.417 ± 0.045) and generated datasets (mAP 50 : 0.245 ± 0.088, mAP 50–95 : 0.144 ± 0.059). Models trained solely on generated images showed significantly lower performance across all categories except the precision metric for red leaf, where performance was comparable to ground truth. This indicated that while AI-generated images can augment datasets and improve generalization, they cannot fully replace ground truth data while maintaining model performance. Integrating AI-generated images with real-world data significantly improved model performance, reduced labor-intensive data collection processes, and provided a more diverse and comprehensive dataset for training, underscoring the importance of a balanced approach to optimizing data collection protocols for wild blueberry cultivation.