<![CDATA[Systematic Literature Review (SLR) is a structured research approach used to map scientific developments, identify research gaps, and provide evidence-based knowledge synthesis. This study aims to systematically review the literature on the application of machine vision, artificial intelligence (AI), and deep learning in egg quality detection, with a particular focus on duck eggs as the research object. Egg quality assessment is crucial in the poultry industry, both to determine suitability for consumption and to ensure successful hatching. However, manual inspection methods are still widely applied, which often result in inaccuracies and inconsistencies. Using the PRISMA methodology, a total of 120 articles published between 2015–2024 were initially identified, of which 45 were selected as relevant studies after screening and eligibility checks. The review results indicate a significant increase in detection accuracy, shifting from conventional image-processing techniques to advanced algorithms such as CNN, ResNet-50, and YOLOv8, achieving accuracies above 94%. Major challenges remain, including the lack of publicly available datasets, risks of overfitting, and limited real-world implementation. This study concludes that future research directions should focus on the integration of lightweight IoT-based systems, standardized duck egg datasets, and hybrid methods (image–spectroscopy) to improve accuracy, robustness, and practical adoption of egg quality detection systems.]]>
Copyrights © 2025
