Classification of handwritten drawing shape in 3D Direct Drawing System using Machine Learning

抄録

The 3D direct drawing system can create the point, the segment, the polygonal and curved surface, and the solid model. However, creating the model is difficult for a user because the command is multiple and complicated. In this paper, the drawing trends and trajectories are investigated. After that, they are predicted and converted to the desired shape using the k-NN method and Neural Network method of machine learning, and the accuracy rates of the two methods are compared.