Bombus Species Image Classification
This addresses a practical challenge for entomologists and ecologists in field work, but it is incremental as it applies existing methods to a new dataset with limited performance gains.
The study tackled the problem of rapid and accurate bumble bee species identification by testing image classification systems using transfer learning, achieving up to 27% accuracy for single species and 50% top-3 accuracy with a composite model.
Entomologists, ecologists and others struggle to rapidly and accurately identify the species of bumble bees they encounter in their field work and research. The current process requires the bees to be mounted, then physically shipped to a taxonomic expert for proper categorization. We investigated whether an image classification system derived from transfer learning can do this task. We used Google Inception, Oxford VGG16 and VGG19 and Microsoft ResNet 50. We found Inception and VGG classifiers were able to make some progress at identifying bumble bee species from the available data, whereas ResNet was not. Individual classifiers achieved accuracies of up to 23% for single species identification and 44% top-3 labels, where a composite model performed better, 27% and 50%. We feel the performance was most hampered by our limited data set of 5,000-plus labeled images of 29 species, with individual species represented by 59 -315 images.