A Multi-Flow Production Line for Sorting of Eggs Using Image Processing

Area of Science:

• Agricultural Engineering
• Computer Vision
• Food Science

Background:

• Traditional egg classification relies on manual methods or load cells, which can be slow and costly.
• Image processing offers a promising alternative for rapid and economical data analysis in the agri-food sector.

Purpose of the Study:

• To develop and evaluate an image processing system for real-time chicken egg classification.
• To assess the feasibility of using visual data for accurate egg weight estimation in industrial settings.

Main Methods:

• A color camera captured images of moving eggs on a motorized roller conveyor within an illumination cabinet.
• An image processing approach was implemented for real-time egg classification and weight prediction.
• Statistical analysis, including linear correlation, was used to validate the results.

Main Results:

• The image processing system successfully classified chicken eggs in real-time on an industrial production line.
• Significant correlations were found between the estimated and actual egg weights.
• A high coefficient of linear correlation (R² = 0.95) was achieved, indicating strong agreement.

Conclusions:

• Image processing provides a viable, efficient, and cost-effective solution for real-time egg classification.
• The developed system demonstrates high accuracy in predicting egg weight, meeting industrial requirements.