EAGER: Artificial Intelligence (AI) to accelerate plant species discovery

植物标本馆是压缩和干燥植物标本的档案科学收藏。这些藏品包含了大量未命名的标本，其中一些对科学来说可能是新的。已知的维管植物大约有40万种，估计还有8万种有待发现，其中许多很可能已经在植物标本馆收藏了。由于气候变化的紧迫威胁，我们需要新的工具来加快这些集合中物种发现的步伐，以便在物种灭绝之前更好地保护它们。联合国的一份报告指出，超过100万种物种面临灭绝的危险，在这一可怕的预测中，最近的一项估计表明，植物的消失速度比动物更快。全世界大约有3000个植物标本馆，它们是植物多样性数据的巨大储存库：这些藏品不仅代表了大量的植物多样性，而且由于植物标本馆收藏的标本可以追溯到几百年前，它们提供了植物多样性随时间变化的快照。植物标本室标本不仅保持了其形态特征，而且还包括了采集日期和地点。这些信息，再加上数以百万计的植物收藏，为大规模了解植物多样性和了解其如何随时间变化提供了框架。人工智能（AI）是一种强大的工具，可以极大地加速植物标本室的物种发现。本项目将为博士后提供培训和职业发展机会。该项目将开发一种工具，利用人工智能快速管理和识别植物标本：这个易于使用的在线工具iCurate将被世界各地的植物标本馆使用，通过输入自己的标本图像，自动识别和管理植物标本馆标本。该工具将从GBIF和iDigBio数据库中广泛存在的数字化维管植物标本中构建，大大减少了识别标本的时间，并允许植物标本馆管理他们缺乏专业知识的植物群。iCurate工具将通过一个逐步优化和扩展的过程，该过程将使用提交给一系列在线竞赛的众包人工智能模型。比赛将使用越来越多的物种和图像进行模型训练，最终包括GBIF和iDigBio中所有公开的维管植物标本图像。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Herbaria are archival scientific collections of pressed and dried plant specimens. These collections contain an overwhelming number of un-named specimens, some of which may be new to science. There are approximately 400,000 known vascular plant species with an estimated 80,000 still to be discovered—many of which are likely already in herbarium collections. With the urgent threats of climate change we need new tools to quicken the pace of species discovery within these collections to better conserve species before they go extinct. A United Nations report indicates that more than one million species are at risk of extinction, and amid this dire prediction a recent estimate suggests plants are disappearing more quickly than animals. There are approximately 3,000 herbaria worldwide and they are massive repositories of plant diversity data: these collections not only represent a vast amount of plant diversity, but since herbarium collections include specimens dating back hundreds of years, they provide snapshots of plant diversity through time. Herbarium specimens not only maintain their morphological features but also include collection dates and locations. This information, multiplied by millions of plant collections, provides the framework for understanding plant diversity on a massive scale and learning how it has changed over time. Artificial Intelligence (AI) is a powerful tool that can vastly accelerate species discovery in herbaria. This project will provide training and professional development opportunities for a post-doctoral researcher.This project will develop a tool for rapid curation and identification of plant specimens from herbaria collections using AI: this easy-to-use online tool, iCurate, will be readily used by herbaria world-wide to automatically identify and curate herbarium specimens by inputting their own specimen images. This tool will be built from the wide range of digitized vascular plant specimens present in the GBIF and iDigBio databases, significantly reducing the time to identify specimens and allowing herbaria to curate plant groups for which they lack expertise. The iCurate tool will be optimized and expanded through a stepwise process that uses crowd-sourced AI models submitted to a series of online competitions. The competitions will use an ever- increasing number of species and images for model training eventually including all public vascular plant specimen images in GBIF and iDigBio.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.