Development of a system for the automated identification of herbarium specimens with high accuracy

Herbarium specimens are dried plants mounted onto paper. They are used by a limited number of researchers, such as plant taxonomists, as a source of information on morphology and distribution. Recently, digitised herbarium specimens have begun to be used in comprehensive research to address broader issues. However, some specimens have been misidentified, and if used, there is a risk of drawing incorrect conclusions. In this study, we successfully developed a system for identifying taxon names with high accuracy using an image recognition system. We developed a system with an accuracy of 96.4% using 500,554 specimen images of 2171 plant taxa (2064 species, 9 subspecies, 88 varieties, and 10 forms in 192 families) that grow in Japan. We clarified where the artificial intelligence is looking to make decisions, and which taxa is being misidentified. As the system can be applied to digitalised images worldwide, it is useful for selecting and correcting misidentified herbarium specimens.

• Publication year

  2022

• Venue

  Scientific Reports

• Publication date

  2022-01-24

• Fields of study

  Biology, Medicine, Environmental Science

• Identifiers
  DOI 10.1038/s41598-022-11450-yPMID 35577859PMCID 9110755
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar, PubMed

• No claims are published for this paper.

• No concepts are published for this paper.

• The Herbarium 2021 Half–Earth Challenge Dataset and Machine Learning Competition

  2022cited by this paper
• The Herbarium 2021 Half-Earth Challenge Dataset

  2021cited by this paper
• Mining museums for historical DNA: advances and challenges in museomics.

  2021cited by this paper
• Improving weeds identification with a repository of agricultural pre-trained deep neural networks

  2020cited by this paper
• Integrating multiple sources of ecological data to unveil macroscale species abundance

  2020cited by this paper
• The Herbarium Challenge 2019 Dataset

  2019cited by this paper
• Plant Seedlings Classification Using Deep Learning

  2019cited by this paper
• Simple but long-lasting: A specimen imaging method applicable for small- and medium-sized herbaria

  2019cited by this paper
• Overview of LifeCLEF Plant Identification Task 2019: diving into Data Deficient Tropical Countries

  2019cited by this paper
• Breaking the silence of the 500-year-old smiling garden of everlasting flowers: The En Tibi book herbarium

  2019cited by this paper
• Automated plant species identification—Trends and future directions

  2018cited by this paper
• Deep learning in agriculture: A survey

  2018cited by this paper
• Automated Identification of Herbarium Specimens at Different Taxonomic Levels

  2018cited by this paper
• Going deeper in the automated identification of Herbarium specimens

  2017cited by this paper
• Retrospective analysis of heavy metal contamination in Rhode Island based on old and new herbarium specimens1

  2017cited by this paper
• How deep learning extracts and learns leaf features for plant classification

  2017cited by this paper
• Digitization of herbaria enables novel research.

  2017cited by this paper
• Mobilizing and integrating big data in studies of spatial and phylogenetic patterns of biodiversity

  2016cited by this paper
• Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization

  2016cited by this paper
• Grad-CAM: Why did you say that? Visual Explanations from Deep Networks via Gradient-based Localization

  2016cited by this paper
• Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning

  2016cited by this paper
• Deep learning for plant identification using vein morphological patterns

  2016cited by this paper
• On natural history collections, digitized and not: a response to Ferro and Flick

  2016cited by this paper
• Plant species classification using deep convolutional neural network

  2016cited by this paper
• Deep Residual Learning for Image Recognition

  2015cited by this paper
• FaceNet: A unified embedding for face recognition and clustering

  2015cited by this paper
• Rethinking the Inception Architecture for Computer Vision

  2015cited by this paper
• Widespread mistaken identity in tropical plant collections.

  2015cited by this paper
• Very Deep Convolutional Networks for Large-Scale Image Recognition

  2014cited by this paper
• Thirteen decades of foliar isotopes indicate declining nitrogen availability in central North American grasslands.

  2010cited by this paper
• Trends and future directions.

  2006cited by this paper
• A Sino-German λ 6 cm polarization survey of the Galactic plane I . Survey strategy and results for the first survey region

  2006cited by this paper
• Herbarium specimens demonstrate earlier flowering times in response to warming in Boston.

  2004cited by this paper

• A survey of neural network segmentation and validation on plant specimen images of Korean Violaceae

  2026cites this paper
• Integrating Deep Convolutional Models and Field Exploration for Wild Allium Species Identification in the Indian Himalayas

  2025cites this paper
• AI-based identification of Chinese vascular plants from herbarium specimens: a tool for all herbaria with Chinese holdings.

  2025cites this paper
• AI for difficult herbarium specimens: identification of peat mosses (subgenus Sphagnum) without dissection.

  2025cites this paper
• Advancing biological taxonomy in the AI era: deep learning applications, challenges, and future directions

  2025cites this paper
• Review of herbarium plant identification of crop wild relatives using convolutional neural network models

  2025influential citation
• A review of artificial intelligence in herbarium specimen image analysis

  2025cites this paper
• Comparative analysis of AlexNet and ResNet50-based classification models for crop wild relatives identification

  2025cites this paper
• Advances in machine learning models for plant species identification: A scoping review

  2025cites this paper
• A pipeline to compile expert‐verified datasets of digitised herbarium specimens for automated plant identification to accelerate taxonomy

  2025cites this paper
• Automated Recognition of Cereal Crop Wild Relatives Using Deep Convolutional Models

  2025cites this paper
• Plant Leaf Identification Using Feature Fusion of Wavelet Scattering Network and CNN With PCA Classifier

  2025cites this paper
• Digitization as scientific labor: Critical approaches and recommendations for herbaria

  2025cites this paper
• The digitisation workflow of the herbarium of the State Museum of Natural History of the NAS of Ukraine (LWS)

  2025cites this paper
• FORMING A LARGE, MULTICENTRE HERBARIUM: HERBARIUM HAUSSKNECHT (JE) JOINS HERBARIUM SENCKENBERGIANUM

  2025cites this paper
• Herbariomic approach solved identity crisis of the putatively extinct Armeria arcuata Welw. ex Boiss. & Reut. (Plumbaginaceae).

  2025cites this paper
• A database for automatic identification of herbarium specimens in Piperaceae family

  2025influential citation
• Nomenclatural updating of the Manitoba Museum Herbarium

  2025cites this paper
• Data Fusion for Integrative Species Identification Using Deep Learning.

  2025cites this paper
• On herbarium specimen images and artificial intelligence.

  2025cites this paper
• Collectomics in plant biodiversity research − looking into the past to understand the present and shape the future

  2025cites this paper
• Exploring the landscape of automated species identification apps: Development, promise, and user appraisal

  2024cites this paper
• Hespi: a pipeline for automatically detecting information from herbarium specimen sheets

  2024cites this paper
• Assembly-free reads accurate identification (AFRAID) approach outperforms other methods of DNA barcoding in the walnut family (Juglandaceae)

  2024cites this paper
• NAFlora-1M: Continental-Scale High-Resolution Fine-Grained Plant Classification Dataset

  2024cites this paper
• Geographic Range vs. Occurrence Records in Plant Distribution Mapping: The Case of Arbutus in the Old World

  2023cites this paper
• Türkiye'nin Bazı Herbaryumlarındaki Salicaceae Koleksiyonu Verilerinin Analizi

  2023cites this paper
• Enhancing Plant Identification: Exploring Deep Learning for Herbarium Specimen Classification

  2023influential citation
• Automated plant species identification from the stomata images using deep neural network: A study of selected mangrove and freshwater swamp forest tree species of Bangladesh

  2023cites this paper
• Species density per grid cell no longer predicts the local abundance of woody plants

  2022cites this paper