Detection and Annotation of Plant Organs from Digitized Herbarium Scans using Deep Learning
This addresses the need for efficient information extraction from herbarium specimens for scientific research, though it is incremental.
The study tackled the problem of automatically detecting plant organs on digitized herbarium scans using Faster R-CNN, achieving strong performance on leaves and stems but less so on flowers.
As herbarium specimens are increasingly becoming digitized and accessible in online repositories, advanced computer vision techniques are being used to extract information from them. The presence of certain plant organs on herbarium sheets is useful information in various scientific contexts and automatic recognition of these organs will help mobilize such information. In our study we use deep learning to detect plant organs on digitized herbarium specimens with Faster R-CNN. For our experiment we manually annotated hundreds of herbarium scans with thousands of bounding boxes for six types of plant organs and used them for training and evaluating the plant organ detection model. The model worked particularly well on leaves and stems, while flowers were also present in large numbers in the sheets, but not equally well recognized.