Cheadle Center for Biodiversity and Ecological Restoration

This research examines the effect of disruptive camouflage by studying the coloration of bees. Using the OpenCV and scikit-learn libraries in Python and incorporating the image theory, I developed a procedure that quantifies and analyzes the visual hotspots on bees, which is mainly applied to identify camouflage effects of different abdominal colors, namely, black and white stripes, black and yellow stripes, and iridescent hues. Patterns that contribute to disruptive camouflage are highlighted by K-mean clustering method. Through the application of the CV methods, we may obtain a quantitative perspective when assessing the qualitative coloration of the bees. Based on image theory-driven clustering, our method provides a foundation that could be extended to further hypothesize the camouflage functionality of coloration patterns.This poster was presented at the UCSB Undergraduate Research Colloquium in 2024.