The aim of this research was to track the growth of chicken eggs, and make a decision as to whether the egg was fertilized or not. A digital imaging system has been developed in order to take an image from six different points without damaging the egg shell. All the images were transferred to a PC and turned into binary images. All the images were reduced to 1024 pixels and fed directly into the classification algorithm. The logistic regression method was used to discriminate the fertility of the eggs. Python programming language and the scikit-learn machine learning library was used to carry out the classifications. True positive, true negative, wrong positive, and wrong negative detection numbers in the trials were 350, 344, 56, and 50, respectively. Negative indicates the egg was infertile, and positive indicated that the egg was fertilized. The model accuracy was measured as 0.8675.