hespi 0.3.0

HErbarium Specimen sheet PIpeline

HErbarium Specimen sheet PIpeline

Hespi takes images of specimen sheets from herbaria and first detects the various components of the sheet. These components include:

• small database label

• handwritten data

• stamp

• annotation label

• scale

• swing tag

• full database label

• database label

• swatch

• institutional label

• number

Then it takes any institutional label and detects the following fields from it:

• family,

• genus,

• species,

• infrasp_taxon,

• authority,

• collector_number,

• collector,

• locality,

• geolocation,

• year,

• month,

• day,

These text fields are then run through the OCR program Tesseract.

Installation

Install hespi using pip:

pip install hespi

The first time it runs, it will download the required model weights from the internet.

It is recommended that you also install Tesseract so that this can be used in the text recognition part of the pipeline.

Usage

To run the pipeline, use the executable hespi and give it any number of images:

hespi image1.jpg image2.jpg

This will prompt you to specify an output directory. You can set the output directory with the command with the --output-dir argument:

hespi images/*.tif --output-dir ./hespi-output

The detected components and text fields will be cropped and stored in the output directory. There will also be a CSV file with the text recognition results for any institutional labels found.

Credits

Robert Turnbull, Karen Thompson, Emily Fitzgerald, Jo Birch.

Publication and citation details to follow.

This pipeline depends on YOLOv5, torchapp, Microsoft’s TrOCR.

Logo derived from artwork by ka reemov.