five

Illustrations of handbook of the birds of the world: Datasets of RGB values and color classification of birds

收藏
NIAID Data Ecosystem2026-05-10 收录
下载链接:
http://datadryad.org/dataset/doi%253A10.5061%252Fdryad.70rxwdc6s
下载链接
链接失效反馈
官方服务:
资源简介:
Plumage color, an important trait of birds, has been widely studied from the perspectives of taxonomy, macroevolution, biogeography, and behavioral mechanisms. However, quantifying avian plumage color remains challenging, and data on the classification and proportion of plumage colors at the species or subspecies level are rarely reported. Some studies have examined plumage colors using illustrations from bird field guides, and hand-drawn illustrations have been shown to be an effective and rich resource for retrieving plumage data in large-scale ecological and biological studies. Here, we collected all illustrations from Handbook of the Birds of the World Alive (HBW Alive) and quantified them as RGB values. HBW Alive includes scientific illustrations of more than 10,000 species and subspecies on a uniform white background, which facilitates plumage quantification. We extracted RGB values from each image, excluding the white background, and applied the K-means clustering algorithm to categorize plumage colors into 24 groups. The complete datasets comprise 18,757 illustrations of 10,290 bird species, with RGB values of 1,000 pixels and the proportion of 24 colors recorded for each image. This dataset includes 5,288 subspecies, male and female plumages of 2,618 species, and more than two morphs for 456 species. Although illustrations have inherent limitations in capturing subtle color variations, this comprehensive dataset provides a valuable resource for investigating ecological, evolutionary, and biological questions across scales, such as species divergence, phenotypic variation in response to climate change and human activities, and global patterns of avian biodiversity. The datasets are not copyright restricted and we encourage researchers to cite this paper when using them in publications.
创建时间:
2025-12-10
二维码
社区交流群
二维码
科研交流群
商业服务