期刊
APPLICATIONS IN PLANT SCIENCES
卷 8, 期 6, 页码 -出版社
WILEY
DOI: 10.1002/aps3.11372
关键词
deep learning; digitized herbarium specimens; Equisetales; ferns; horsetails; machine learning
Premise Equisetumis a distinctive vascular plant genus with 15 extant species worldwide. Species identification is complicated by morphological plasticity and frequent hybridization events, leading to a disproportionately high number of misidentified specimens. These may be correctly identified by applying appropriate computer vision tools. Methods We hypothesize that aerial stem nodes can provide enough information to distinguish amongEquisetum hyemale,E. laevigatum, andE. xferrissii, the latter being a hybrid between the other two. An object detector was trained to find nodes on a given image and to distinguishE. hyemalenodes from those ofE. laevigatum. A classifier then took statistics from the detection results and classified the given image into one of the three taxa. Both detector and classifier were trained and tested on expert manually annotated images. Results In our exploratory test set of 30 images, our detector/classifier combination identified all 10E. laevigatumimages correctly, as well as nine out of 10E. hyemaleimages, and eight out of 10E. xferrissiiimages, for a 90% classification accuracy. Discussion Our results support the notion that computer vision may help with the identification of herbarium specimens once enough manual annotations become available.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据