Work Description

Date: 21 March, 2025

Dataset Title: Passerine Skeleton Images

Dataset Contact: Brian C. Weeks [email protected]

Dataset Creator:
Name: Brian C. Weeks
Email: [email protected]
Institution: University of Michigan Museum of Zoology

Funding: The David and Lucile Packard Foundation

Key Points:
- We provide images of museum skeletal specimens that were used to identify and measure 12 skeletal elements using an automated computer vision approach.

Research Overview:
Functional traits are commonly used to understand the evolutionary history of taxa and to examine ecological dynamics at large scales. Birds are rapidly becoming a model system for this type of macroevolutionary and macroecological work, and this dataset adds to existing datasets on external morphological traits, distributional data, phylogenetic data and ecological data. These images have been used in an effort to measure skeletal traits in an effort to integrate musculoskeletal traits into our understanding of bird macroevolution and macro ecology.

Methodology:
For each image, the specimen was randomly placed on the surface, the skull and keel were oriented so that their profiles were captured in the image, the associated specimen tag was placed so that the catalog number was visible, and the remaining bones were spread haphazardly across the surface using a paintbrush. Photographs were then taken of the specimen following the methodology described in: Weeks et al. 2023. A deep neural network for high-throughput measurement of functional traits on museum skeletal specimens. Methods in Ecology and Evolution 14(2): 347-359. doi: https://doi.org/10.1111/2041-210X.13864.

Files contained here:
High Resolution RGB Photographs

Each folder is named according to the University of Michigan Museum of Zoology specimen that was photographed. Within each folder there is a .tiff file. These files have been saved with the UMMZ catalog number in the filename (skeleton-UMMZCATALOG#-Color2.tiff). The files are high resolution (5,472 x 3,468 pixels) images of the specimen captured with the FLIR Blackfly S camera (with a SONY IX183 sensor). These images were used to generate the trait data presented in: Weeks et al. 2024. Skeletal trait measurements for thousands of bird species. bioRxiv. doi: https://doi.org/10.1101.2024/12/19/629481.

Stereoscopic Depth Camera RGB File

Within each folder there is a Color.png file. These files have been saved with the UMMZ catalog number in the filename (skeleton-UMMZCATALOG#-Color.png). The files are the RGB images (1,280 x 720 pixels) captured by an Intel RealSense stereoscopic camera as outlined in: Weeks et al. 2023. A deep neural network for high-throughput measurement of functional traits on museum skeletal specimens. Methods in Ecology and Evolution 14(2): 347-359. doi: https://doi.org/10.1111/2041-210X.13864. These files were not used in the Skelevision model, and have not been used to generate trait data, but have been retained for future use.

Stereoscopic Camera Depth Image

Within each folder there is a Depth.png file. These files have been saved with the UMMZ catalog number in the filename (skeleton-UMMZCATALOGX-Depth.png). The files are an RGB image of distance-to-surface data generated by the Intel RealSense camera as outlined in Weeks et al. 2023. A deep neural network for high-throughput measurement of functional traits on museum skeletal specimens. Methods in Ecology and Evolution 14(2): 347-359. doi: https://doi.org/10.1111/2041-210X.13864. These files were not used in the Skelevision model, and have not been used to generate trait data, but have been retained for future use as-needed.

Stereoscopic Depth Camera Data

Within each folder there is a.csv file. These files have been saved with the UMMZ catalog number in the filename (skeleton-UMMZCATALOG#-Depth-data.csv). The files contain depth data representing the distance from the camera to the surface for each pixel in the scene captured by the Intel RealSense Camera (corresponding -Depth.png and -Color.png files) as outlined in Weeks et al. 2023. A deep neural network for high-throughput measurement of functional traits on museum skeletal specimens. Methods in Ecology and Evolution 14(2): 347-359. doi: https://doi.org/10.1111/2041-210X.13864. These data are not absolute values of distance, but can be converted to 3D data, in combination with the RGB image files generated for each specimen. These files were not used in the Skelevision model or to generate trait data, but have been retained for future use as-needed.

Use and Access:
This data is made available under the UMMZ Digital Data Usage Agreement.

To Cite Data:

Brian C. Weeks. University of Michigan Museum of Zoology Passerine Bird Skeleton Images [Data set], University of Michigan - Deep Blue Data.