Abstract:
Eggs are a commonly consumed food item and are essential for daily life. However,
cracks in the eggs can lower their nutritional value and affect their potential. In order
to distinguish between healthy and unhealthy eggs, the authors of thisstudy focused
on creating an automated egg crack classification system using convolutional neural
networks(CNN).
The technique makes use of an eggshell dataset with different types of cracks,such as
small, heavy, and no cracks. When analyzing an image's characteristics and
categorizing it as healthy or unhealthy, the CNN architecture is helpful. For the
purposesofbuildingandassessingmodels,thedatasetissplitintotrainingandtest sets.
The experimental results show how precisely the CNN algorithm classifies egg
damage. With an overall accuracy of 80%, the trained CNN algorithm performs
encouragingly in differentiating between healthy and cracked eggs. The CNN
algorithm's capacity to detect damaged eggs can assist in ensuring egg quality and
reducing the health risks associated with eating cracked eggs.
With less need for manual inspection and more productivity, the suggested system
provides a workable and effective solution for evaluating egg quality in the food
industry. It also lays the groundwork for future investigations into the creation of
automated systems for monitoring egg quality.
