Teeth is a structure in which many vertebrates exist. For some animals, such as lions, tigers and so on, teeth are chewing tools and weapons to protect themselves. But for human, it also carries the beauty of the face. When the teeth are sick, accurate classification of the teeth seems particularly important. The main purpose of this paper is to classify the teeth accurately using biogeography-based optimization algorithm (BBO) and Multilayer perceptron (MLP). The results showed our method achieved 83.75± 2.95%, 83.50± 5.16%, 84.00± 5.16%, and 84.75± 3.43% accuracy for identifying incisor, canine, premolar, and molar.