Automatic Visual Detection of Fresh Poultry Egg Quality Inspection using Image Processing

Poultry production is depending on the excellent quality of eggs. Commercially produced eggs are graded concurrently for exterior quality factors like cleanliness and soundness of the shell and inner quality factors like that of the white, yolk, and size of the air cell before being provided to the market in some industrialized countries. However, systems for determining and grading egg quality are still not fully deployed globally at the expected rate. In the Philippines, this method is still challenging for small to medium poultry farms. Owners still practice the traditional method of checking the internal and external quality of eggs which results in time-consuming, requires more manpower, and issues in validation. This paper introduced the design of automatic visual inspection of chicken fresh egg quality both for internal and external quality. Raspberry-pi was employed for a quick and effective output for image processing approach. Different criteria were considered based on the gathered data from the poultry farm owners in developing the system such as the amount of luminance to be used, the safety of egg examination, and accuracy. Different algorithms were employed to perform the internal and external inspections. Based on test results, the camera detection gathered an average of 77.92mm in ten (10) tests before finding the optimal distance for egg detection and detects an average of 75.94 for accuracy. Furthermore, the variance and standard deviation of the data indicate that the values are centered around the mean. Based on the ten (10) sampled tests made, it generated 95% accuracy of correctly detected fresh poultry chicken eggs quality.