Research On Craniofacial Restoration Method Based On Laplace With Angle Constraint

Craniofacial reconstruction is a technology that is based on statistical prior knowledge to reconstruct the unknown skull face.It has important applications in archeology and forensic science.The method of craniofacial reconstruction based on template deformation has become a hot topic in the study of craniofacial morphology and craniofacial recovery,and it is more scientific and objective than the traditional craniofacial reconstruction.A three dimensional skull registration automatically based on partition and improved Iterated Closest Point(ICP)is presented in this paper.It is used to select reference skull collection,which is similar to the skull to be restored.On the basis of manual calibration of the skull feature points to be restored,the corresponding point matching method was used to automatically calibrate the reference skull 's feature points.The Laplacian algorithm with angle constraint is proposed to achieve craniofacial restoration.The similarity matching of multiple geometric constraints is performed to evaluate the accuracy of the restoration method in this paper.The main research work includes: 1.skull selection algorithms based on partition and improved Iterated Closest Point.Firstly,the craniofacial data which is similar to the anthropological properties of the skull to be restored are found from the craniofacial model database;Secondly,an improved ICP algorithm based on skull partitioning and dynamic estimation(Destimation)is proposed,and it register the skull to be restored and the similar skull.Root mean square error(RMSE)was used as registration error to calculate the error between skull and skull to be restored.According to this algorithm,k similar skulls with minimum registration error are selected as reference skulls,and complete the selection of reference skulls and facial skulls.Experiments have shown that this method can select craniofacial data that are closer to the skull to be restored.2.Research on Craniofacial Restoration Method based on Laplace with Angle Constraint.Firstly,according to the error between the skull to be restored and the reference skull,the improved Laplace is used to solve the feature points of the k reference skins after deformation;Secondly,the weights of the k deformed faces are inversely proportional to the RMSE;finally,the weighted average deformation of each reference face is calcculated to restore skull to be restored.Experiments show that restoration result is more close to real face.3.Evaluating effect of craniofacial restoration which is based on multiple geometric constraints.Firstly,according to the curvature factor and shape factor,selecting the initial matching pair;Secondly,the combination of vertex normals and curvature weighted distance is proposed to effectively eliminate the mismatched corresponding points;finally,the energy function is introduced to obtain the optimal transformation matrix,and evaluate similarity of the restored face and real face.Experiments show that this method can effectively evaluate the results of the recovery method.