Selective-OCR protocol for mass digitization of herbarium specimen labels
DOI:
https://doi.org/10.22201/ib.20078706e.2026.97.5781Keywords:
Biodiversity informatics, Biological collections, Databases, Image processing, MEXUAbstract
It still remains to digitize label data of a high percentage of specimens of vascular plants in the herbariums. There are experiences of the use of the OCR technique to support the process of digitization of specimens, however, it is still necessary to explore and describe in greater detail its limitations and strengths. The digitization of some of the data of the labels from herbarium specimens can be done massively and automatically by applying optical character recognition techniques (OCR). Five target fields (geographic and taxonomic super-primary fields) were selected to obtain their information through OCR applied to 8,451 images of herbarium specimens, guided by a label information architecture and with human intervention. The information contained in the 5 target fields was identified in 70.6% of the labels, 23.7% in 1–4 of the target fields, and only in 5.7% none could be identified. Mistakes ranged from 0.8 to 3.3% depending on the field. OCR cannot automatically identify all information fields of herbarium specimen labels; however, it is possible in a high percentage to retrieve the information of the most consulted fields.
References
Blagoderov, V., Kitching, I. J., Livermore, L., Simonsen, T. J., & Smith, V. S. (2012). No specimen left behind: industrial scale digitization of natural history collections. Zookeys, 209, 133–146. https://doi:10.3897/zookeys.209.3178
Chapman, A. D., Belbin, L., Zermoglio, P. F., Wieczorek, J., Morris, P. J., Nicholls, M. et al. (2020). Developing standards for improved data quality and for selecting fit for use biodiversity data. Biodiversity Information Science and Standards, 4, e50889. https://doi.org/10.3897/biss.4.50889
Conabio (Comisión Nacional para el Conocimiento y Uso de la Biodiversidad). (2019). Datos primarios de ejemplares del Sistema Nacional sobre Biodiversidad (SNIB) –características y reglas –. Ciudad de México. Last access June 3, 2025. www.snib.mx/ejemplares/docs/CONABIO-SNIB-ProtocoloCalidadI.pdf
Corlett, R. T. (2022). We need to accelerate the digitization of existing botanical information and complete the global plant inventory. Integrative Conservation, 1, 6–7. https://doi.org/10.1002/inc3.12
Friedl, J. E. (2006). Mastering regular expressions. Sebastopol, CA: O'Reilly Media, Inc.
Guralnick, R., LaFrance, R., Denslow, M., Blickhan, S., Bouslog, S. M., Yost, J. et al. (2024). Humans in the loop: community science and machine learning synergies for overcoming herbarium digitization bottlenecks. Applications in Plant Sciences, 12, e11560. https://doi.org/10.1002/aps3.11560
Kirchhoff, A., Bügel, U., & Santamari, E. (2018). Toward a service-based workflow for automated information extraction from herbarium specimens. Database, 2018, bay103. https://doi.org/10.1093/database/bay103
Korzynski, P., Mazurek, G., Krzypkowska, P., & Kurasinski, A. (2023). Artificial intelligence prompt engineering as a new digital competence: analysis of generative AI technologies such as ChatGPT. Entrepreneurial Business and Economics Review, 11, 25–37. https://doi.org/10.15678/EBER.2023.110302
Morville, P., & Rosenfeld, L. (2006) Information architecture for the World Wide Web: designing large-scale web sites. Sebastopol, CA: O’Reilly Media, Inc.
Murguía-Romero, M., Ortiz, E., Serrano-Estrada, B., & Villaseñor, J. L. (2023). The Kew’s" World Checklist of Vascular Plants" and its relevance to the knowledge of the flora of Mexico. Botanical Sciences, 101, 632–653. https://doi.org/10.17129/botsci.3223
Murguía-Romero, M., Serrano-Estrada, B., Salazar, G., Gernandt, D. S., Melo-Samper-Palacios, U., Sánchez-González, G. E. et al. (2023). IBdata v3 “Helia Bravo Hollis”. Manual de uso. Ciudad de México: Instituto de Biología, Universidad Nacional Autónoma de México.
Murguía-Romero, M., Serrano-Estrada, B., Salazar, G. A., Sánchez-González, G. E., Melo-Samper-Palacios, U., Gernandt, D. S. et al. (2024). The IBdata web system for biological collections: design focused on usability. Biodiversity Informatics, 18, 1–12. https://doi.org/10.17161/bi.v18i.20516
Nelson, G., Sweeney, P., Wallace, L. E., Rabeler, R. K., Allard, D., Brown, H., et al. (2015). Digitization workflows for flat sheets and packets of plants, algae, and fungi. Applications in Plant Sciences, 3, 1500065. https://doi.org/10.3732/apps.1500065
Pavlick, G. (2025). What is Generative AI (GenAI)? How does it work? Oracle. Accessed June 6, 2025: https://www.oracle.com/uk/artificial-intelligence/generative-ai/what-is-generative-ai/
Schank, R. C., & Abelson, R. P. (1975). Scripts, plans, and knowledge. In IJCAI’75: Proceedings of the 4th International Joint Conference on Artificial Intelligence, Volume 1 (pp. 151–157). San Francisco, CA: Morgan Kaufmann Publishers Inc. https://dl.acm.org/doi/10.5555/1624626.1624649
Schank, R. C., & Abelson, R. P. (1977). Scripts, plans, goals, and understanding: an inquiry into human knowledge structures. Hillsdale, New Jersey: Lawrence Eribaum Associates Inc. Publishers.
Shanahan, M. (2024). Talking about large language models. Communications of the ACM, 67, 68–79. https://doi.org/10.1145/3624724
Soltis, P. S., Nelson, G., & James, S. A. (2018). Green digitization: online botanical collections data answering real-world questions. Applications in Plant Sciences, 6, e1028. https://doi.org/10.1002/aps3.1028
Takano, A., Cole, T. C., & Konagai, H. (2024). A novel automated label data extraction and database generation system from herbarium specimen images using OCR and NER. Scientific Reports, 14, 112. https://doi.org/10.1038/s41598-023-50179-0
