Predicting age is one of the most important characteristics in the forensic context for human identification because biological profiles can only be effectively rebuilt by estimating the most precise age. Humans usually count their age using calendars that are pre-programmed to keep track of days. It is not difficult for any living human being to measure their age, but it becomes a challenge when forensics needs to determine the exact age of a dead person involved in a criminal case. By comparing the stage of tooth development in X-rays to established dental growth norms, forensic scientists may estimate age. To deal with such problems, an age estimate system that predicts a person's real age group is required. In this paper, Convolutional Neural Network (CNN) was used to build Human Age Estimation System that can predict a person's actual age group. The system employs Inception-v3 architecture to create the model. The dataset consists of 1404 dental X-ray images of pre-teen in Malaysia. The model was trained using 60% of the dataset, and the remaining 40% was used for testing. Accuracy, precision, recall, and f-measure were used to assess the model. The model achieved 82%, 78%, 76%, and 77%, respectively. As a conclusion, a large amount of data is required for the training and testing process to acquaint the model with the pattern of tooth growth by age group. It is also crucial to rule out any illnesses that impact tooth growth, as well as radiographic images that are not of good quality.
