A cost-effective approach for chicken egg weight estimation through computer vision

Authors : Alphany ARAGUA, Val İrvin MABAYO

Pages : 82-87

Doi:10.31015/jaefs.18013

View : 10 | Download : 7

Publication Date : 2018-09-20

Article Type : Research Paper

Abstract :Egg weighing and classification are among the most significant phases done in egg processing by industries which are tedious if done manually by poultry owners, and egg inspectors and graders.  This study presented an alternative way of estimating chicken egg weight through computer vision minimizing human interaction during the process. In this study, fifteen eggs of white leghorn chicken layers of different sizes were tested. The eggs’ image was captured using an inexpensive yet reliable webcam which was then loaded onto the MatLab workspace for image processing and further image analysis. The center of gravity of the image was determined, and the extraction of minor axis length and major axis length followed. The obtained values were used to compute the egg’s weight mathematically. Through the different image processing methods, image dimensions were extracted and used to calculate the desired output. The results of this study showed 96.31% accuracy in estimating the egg’s weight and classification validated by manual egg weighing and classification procedure. 
Keywords : Egg weight, webcam, image processing, image analysis