In digital orthodontic field, obtaining complete crown mesh is a very important part. Due to the tightness of adjacent teeth and the connection of teeth to the gums, both sides of the crown data obtained after scanning and segmentation are usually missing, and repairing the crown mesh data is a critical task. There are many researchers worked on dental mesh repairing problem, although their methods could obtain high quality results, they do not consider about tooth shape and usually require user interaction. In this paper, we propose a dental mesh repairing method based on statistical shape analysis. We estimate the shape parameters such that the tooth shape is as close as possible to the boundary region of the crown mesh to be repaired. Then, these parameters are used as a template to perform surface repair on both sides of the crown model. The method proposed in this paper can accurately estimate a set of shape parameters that are close to the input crown model, and the stitch results are smoother. On the other hand, this method does not require human interaction and additional data support, the whole process is automatic.