Toward a service-based workflow for automated information extraction from herbarium specimens

Abstract

Over the past years, herbarium collections worldwide have started to digitize millions of specimens on an industrial scale. Although the imaging costs are steadily falling, capturing the accompanying label information is still predominantly done manually and develops into the principal cost factor. In order to streamline the process of capturing herbarium specimen metadata, we specified a formal extensible workflow integrating a wide range of automated specimen image analysis services. We implemented the workflow on the basis of OpenRefine together with a plugin for handling service calls and responses. The evolving system presently covers the generation of optical character recognition (OCR) from specimen images, the identification of regions of interest in images and the extraction of meaningful information items from OCR. These implementations were developed as part of the Deutsche Forschungsgemeinschaft-funded a standardised and optimised process for data acquisition from digital images of herbarium specimens (StanDAP-Herb) Project.

Using big data to improve ecotype matching for Magnolias in urban forestry

“…Gathering the records for the plant occurrence data was a long-winded process, with each of the large databases presenting their own challenges: GBIF, for example, holds relatively few verified observations of naturally occurring Magnolia populations but offers excellent data transfer capabilities, Global Plants hosts a large number of records but makes data transfer challenging, whilst the Chinese Virtual Herbarium requires translation from Mandarin and an iterative process of positive identification and filtering to derive accurate records. The archival research in herbaria was highly effective but corroborated reports that only a small fraction of plant records are hosted by online databases (Harris and Marsico, 2017;Kirchhoff et al, 2018), and as a result future applications of this methodology should factor in the extensive desktop research. By contrast, climate data was straightforward to derive, with the principal short-comings being the grain and accuracy of the data, especially when assessing urban environments in comparison to their rural hinterlands.…”

Reversing extinction trends: new uses of (old) herbarium specimens to accelerate conservation action on threatened species

“…Furthermore, machine learning technologies (e.g. OCR, optical character recognition) can be applied to the extraction of metadata from herbarium specimens, for example, collectors names (Silva, 2016), plant traits from textual descriptions (Dagtekin et al ., 2018), and label content (Heidorn & Wei, 2008; Kirchhoff et al ., 2018; Walton et al ., 2020), enhancing integrative and cross‐linked research on complex phenomena. Tiered data gathering and analysis from – to date – unconnected data expand the information that herbarium collections can provide (Soltis et al ., 2018; Lendemer et al ., 2019; Theeten et al ., 2019).…”

Identification of herbarium specimen sheet components from high‐resolution images using deep learning

“…not actually implemented) (e.g. Haston et al, 2015 ; Kirchhoff et al, 2018 ; Moen et al, 2010 ). Some investigations (e.g.…”