Detecting fertile eggs during incubation using dielectric properties

Separation of fertile and infertile eggs is a major concern of the poultry industry. If this separation is done at the early days of the incubation period, less space will be occupied in the incubator resulting in improved capacity and efficiency. Dielectric constant and loss factor were used in this study to detect infertile Hy-line eggs. In this study, 126 eggs were selected and their dielectric properties were measured on the third day of incubation. On the fifth day, it was detected that there were 112 fertile and 14 infertile eggs. Data were then analyzed by Weka software using artificial neural network (ANN) and support vector machine (SVM) classifiers. Results showed that, on the third day of incubation, SVM and ANN successfully detected infertile and fertile eggs with 100% accuracy. Detecting the fertile egg by using electrical conductivity is a non-destructive method and is also more accurate than image processing, signal processing and PCR methods.