There are many techniques used for diagnosing diseases and identifying a person, and the iris of the eye helps in identifying a person, and the colour of the eye helps in identifying some diseases. People with coloured eyes are more susceptible than others to contracting some diseases. In this research, we try to obtain the hue ratio of people’s iris colours from the basic colours of the iris (black, green, and blue). Thus, we proposed a model using a CNN network, which is one of the deep learning algorithms, to identify the colour percentage of the eye, as determining the colour of the eye contributes to the diagnosis of some diseases such as skin cancer. The dataset from Roboflow was used to train the model, and real images were used to test the proposed model. The proposed CNN model built using the Python programming language is explained. We obtained 98.85% accuracy for training the model, while the accuracy for testing the model was 98. 44%.