当前位置: X-MOL 学术Biodivers. Data J. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Detection and annotation of plant organs from digitised herbarium scans using deep learning
Biodiversity Data Journal ( IF 1.0 ) Pub Date : 2020-12-10 , DOI: 10.3897/bdj.8.e57090
Sohaib Younis 1, 2 , Marco Schmidt 1, 3 , Claus Weiland 1 , Stefan Dressler 4 , Bernhard Seeger 2 , Thomas Hickler 1
Affiliation  

As herbarium specimens are increasingly becoming digitised 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 mobilise such information. In our study, we use deep learning to detect plant organs on digitised 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 were not equally well recognised.

中文翻译:


使用深度学习从数字化植物标本馆扫描中检测和注释植物器官



随着植物标本馆标本日益数字化并可在在线存储库中访问,先进的计算机视觉技术被用来从中提取信息。植物标本馆表上某些植物器官的存在在各种科学背景下都是有用的信息,并且这些器官的自动识别将有助于调动这些信息。在我们的研究中,我们使用深度学习通过 Faster R-CNN 检测数字化植物标本上的植物器官。在我们的实验中,我们手动注释了数百个植物标本馆扫描,其中包含六种植物器官的数千个边界框,并将它们用于训练和评估植物器官检测模型。该模型在叶子和茎上效果特别好,而纸张中也大量存在花朵,但没有得到同样好的识别。
更新日期:2020-12-10
down
wechat
bug