Our dataset consists of color data extracted from ~50,000 images of ant heads acquired from the database AntWeb.org. The .csv file contains the pixel-mean RGB and HSL values of image segments extracted from the ant head images, in addition to the existing metadata for these 50,000 specimens.

The data was collected from AntWeb.org, a database containing records and metadata for ant specimens housed in museums. Around ~50,000 of these specimens have images of their heads included in the database. Our dataset consists of color data extracted from these ant head images attached to the existing metadata from these 50,000 specimens. This color data consists of the mean RGB and HSL values of all pixels of image segments extracted from the ant head images. Segments were obtained using a DeepLabV3 segmentation model trained with transfer learning on 50 ant head images - the training set and training and inference code can be found in the project folder on GitHub.

The data is in .csv format and can be opened using a spreadsheet editor like Excel or a statistical environment like RStudio.