WingAnalogy: a computer vision-based tool for automated insect wing asymmetry and morphometry analysis.

Area of Science:

• Entomology
• Bioinformatics
• Computational Biology

Background:

• Insect wing morphology is crucial for taxonomy and evolutionary studies.
• Quantifying wing asymmetry provides insights into developmental stability and environmental stress.
• Automated analysis tools are needed to handle large datasets and complex measurements.

Purpose of the Study:

• To introduce WingAnalogy, a novel computer tool for automated insect wing morphology and asymmetry analysis.
• To provide a comprehensive platform for image processing, metric extraction, and data visualization of insect wings.
• To facilitate detailed, localized comparisons of wing structures and asymmetry patterns.

Main Methods:

• Image processing and computer vision for wing segmentation and feature extraction (boundaries, junctions).
• Quantification of wing and cell metrics (area, length, width, circularity, centroid positions).
• Image scaling and superimposition using Particle Swarm Optimization (PSO); calculation of regression and Normalized Root Mean Square Error (NRMSE).

Main Results:

• Successful segmentation and extraction of insect wing structures and cell features.
• Accurate quantification of various morphological and asymmetry parameters.
• Generation of visualizations and detailed reports (PDF, CSV, TXT) on wing asymmetry.

Conclusions:

• WingAnalogy offers a robust and efficient solution for automated insect wing analysis.
• The software enables precise quantification and interpretation of wing morphology and asymmetry.
• WingAnalogy supports localized comparisons by dividing wings into distinct cell sets, enhancing research capabilities.