Classification of Eggs Based on the Internal Quality Using of Computer Vision

Nowadays, concern about egg quality is growing steadily. Freshness makes a major contribution to the quality of egg and egg products. In this regard, in order to achieve both high efficiency and high productivity, eggs’ quality assessment is necessary to enhance safety and quality assurance. Therefore, the aim of this study was to design and fabricate a system to determine the internal quality of eggs using of machine vision. For this purpose, 210 eggs for 30 days were storted at room temperature and humidity in a completely randomized design and some quality parameters of eggs were measured (e.g. total weight of egg, yolk, albumen and eggshell weight, strength and thickness of the eggshell, Haugh unit as destructive parameters and air cell area, and area index (D) as non-destructive parameters). According to statistical analysis total weight of egg, strength and thickness of the shell, yolk and shell weight and yolk color as egg quality parameters were not affected by the storage time (p>0.05). But, the air cell area and the parameter D as non-destructive parameters, was increased significantly at the level of 5%. The results showed Haugh unit was decreased from 9.28 to 2.04 as storage time increased (p<0.05). Classification accuracy using non-destructive data with Parzen method was calculated 84.17%. This shows the ability of the proposed system to classify fresh eggs from unfreash ones with high accuracy.