Optical Character Recognition (OCR) is the use of computer algorithms to detect text information from images. If OCR tools are activated in your Symbiota portal, you can use OCR to attempt to parse textual information from pictures of, e.g., specimen labels and enter data into the appropriate fields from there. Instructions for using batch OCR processes are provided below. For instructions for using OCR in the Occurrence editor, see the OCR page in the Editor Guide.
To run OCR on many specimen images at a time
A video tutorial of this process is shown here:
Some portals also include an interface into which you can upload OCR text files generated outside of the portal environment. For instance, ABBYY FineReader has the ability to batch OCR specimen images and output the results as separate text files (.txt) named after the source image. OCR text files are linked to specimen records by matching catalog numbers extracted from the file name and comparing OCR and iamge file names.
This tool can also be found on the OCR tab in the Processing Toolbox.
To upload a file, make sure your collection and files meet the following requirements:
Katie Pearson. Batch OCR. In: Symbiota Support Hub (2021). Symbiota Documentation. https://biokic.github.io/symbiota-docs/coll_manager/edit/ocr/. Created on 03 Dec 2021.