Microscopy Image Analysis of Butterfly Wing and Moth Eyes using Statistical Feature Learning

This paper shows exploratory analysis of microscopic images of butterfly wings and moth eyes in solar energy applications and biomaterial characterisation. We used handcrafted and automatically extracted statistical features from the scanning electron microscope (SEM) images to train different classification algorithms. Our analysis shows good classification performance in terms of accuracy, sensitivity, and specificity using feature selection, dimensionality reduction, and using deep learning models to analyse microscopic images of biological structures of butterfly wings and moth eyes. Novel material design inspired by biomimicry needs deeper understanding of microscopic image features, which this paper aims to achieve.