Abstract
Modern incubation technology requires precise and objective selection of hatching goose eggs to ensure high hatchability and genetic quality of offspring. Traditional manual sorting methods—based on weighing, visual inspection, and ovoscopy—remain labor-intensive, subjective, and prone to human error. In this study, an intelligent optoelectronic system was developed and experimentally validated for the non-destructive analysis and classification of goose eggs according to their incubation suitability. The system integrates a strain gauge, a digital RGB camera, and an infrared rangefinder, managed by a Raspberry Pi 4 microcomputer running Python software with image-processing algorithms. The research was conducted using 240 incubation eggs of the Lindovskaya goose breed, characterized by large geometric dimensions (average length 90.2 mm, width 68.7 mm, mass 167.3 g). The algorithm calculates morphometric and physical indicators—mass, volume, density, shape index, and shape coefficient (K₁)—to form an objective classification criterion. The results demonstrated a high correlation between manual and automated measurements (r > 0.96) and an average relative error below 2.3%. The proposed system ensures accurate, reproducible, and rapid sorting of hatching goose eggs, significantly reducing human-factor variability. Adaptation of optical and algorithmic parameters, including a correction factor Kᵥ = 0.645 optimized for goose egg geometry, allows the method to be effectively applied in hatcheries and smart-farming systems.