Search for: All records

Abstract Pollen identification is necessary for several subfields of geology, ecology, and evolutionary biology. However, the existing methods for pollen identification are laborious, time-consuming, and require highly skilled scientists. Therefore, there is a pressing need for an automated and accurate system for pollen identification, which can be beneficial for both basic research and applied issues such as identifying airborne allergens. In this study, we propose a deep learning (DL) approach to classify pollen grains in the Great Basin Desert, Nevada, USA. Our dataset consisted of 10,000 images of 40 pollen species. To mitigate the limitations imposed by the small volume of our training dataset, we conducted an in-depth comparative analysis of numerous pre-trained Convolutional Neural Network (CNN) architectures utilizing transfer learning methodologies. Simultaneously, we developed and incorporated an innovative CNN model, serving to augment our exploration and optimization of data modeling strategies. We applied different architectures of well-known pre-trained deep CNN models, including AlexNet, VGG-16, MobileNet-V2, ResNet (18, 34, and 50, 101), ResNeSt (50, 101), SE-ResNeXt, and Vision Transformer (ViT), to uncover the most promising modeling approach for the classification of pollen grains in the Great Basin. To evaluate the performance of the pre-trained deep CNN models, we measured accuracy, precision, F1-Score, and recall. Our results showed that the ResNeSt-110 model achieved the best performance, with an accuracy of 97.24%, precision of 97.89%, F1-Score of 96.86%, and recall of 97.13%. Our results also revealed that transfer learning models can deliver better and faster image classification results compared to traditional CNN models built from scratch. The proposed method can potentially benefit various fields that rely on efficient pollen identification. This study demonstrates that DL approaches can improve the accuracy and efficiency of pollen identification, and it provides a foundation for further research in the field. 

Research on plant-pollinator interactions requires a diversity of perspectives and approaches, and documenting changing pollinator-plant interactions due to declining insect diversity and climate change is especially challenging. Natural history collections are increasingly important for such research and can provide ecological information across broad spatial and temporal scales. Here, we describe novel approaches that integrate museum specimens from insect and plant collections with field observations to quantify pollen networks over large spatial and temporal gradients. We present methodological strategies for evaluating insect-pollen network parameters based on pollen collected from museum insect specimens. These methods provide insight into spatial and temporal variation in pollen-insect interactions and complement other approaches to studying pollination, such as pollinator observation networks and flower enclosure experiments. We present example data from butterfly pollen networks over the past century in the Great Basin Desert and Sierra Nevada Mountains, United States. Complementary to these approaches, we describe rapid pollen identification methods that can increase speed and accuracy of taxonomic determinations, using pollen grains collected from herbarium specimens. As an example, we describe a convolutional neural network (CNN) to automate identification of pollen. We extracted images of pollen grains from 21 common species from herbarium specimens at the University of Nevada Reno (RENO). The CNN model achieved exceptional accuracy of identification, with a correct classification rate of 98.8%. These and similar approaches can transform the way we estimate pollination network parameters and greatly change inferences from existing networks, which have exploded over the past few decades. These techniques also allow us to address critical ecological questions related to mutualistic networks, community ecology, and conservation biology. Museum collections remain a bountiful source of data for biodiversity science and understanding global change.