Abstract. This paper proposed an interactive tooth segmentation method based on Geodesic which aims to solve the inaccurate problem of segmentation in dental model for irregular teeth case. The features of the teeth are automatically acquired according to the prior knowledge. Then based on those features, the boundary of each tooth is obtained by the improved regional growth method and corrected by combining the correction points from the user inputting based on the shortest geodesic path, so that, the teeth are precisely segmented according to the boundary. The experimental results show that our method can be used to segment different kinds of teeth correctly and the computational speed and the precision of segmentation are improved obviously. Furthermore, as the features and the initial boundaries can be acquired automatically, the complexity of the implementation is significantly reduced.

Generally, the performance of deep learning models is related to the captured features of training samples. When the training samples belong to different domains, the diverse features may increase the difficulty of training high performance models. In this paper, we built a new framework that generates multiple models on the organized samples to increase the accuracy of classification. Firstly, our framework selects some existing models and trains each of them on organized training sets to get multiple trained models. Secondly, we select some of them based on a validation set. Finally, we use some fusion method on the outputs of the selected models to get more accurate results. The experimental results show that our framework achieved higher accuracy than the existing methods. Our framework can be an option for the deep learning system to increase the classification accuracy.