Mejora de las fronteras en mallas simpliciales de múltiples materiales generadas a partir de imágenes

Abstract

A digital image may contain objects made up of multiple regions concerning different material properties, physical or chemiqcal attributes. Thus, generating multi-material simplicial meshes (of 2 and 3 dimension) from images may produce singularities on boundaries between partitioned regions. Most of current methods remove singularities considering only meshes representing a single material or object. This work focuses on reparing singularities of meshes describing multiple objects. For this purpose, our approach is based on partitioning adjustment (relabeling) and point insertion, these ideas can be combined to improve different qualities: increase the speed, keep as much as posible the shape of the original partitioning or keep the number elements in the mesh. Our proposal was applied in different contexts, meshes with different number of singularities or materials, of 2 and 3 dimensions. In all cases, all the singularities were removed from the mesh. Addicionally, once the mesh has no singularities, we could develop and apply some alternatives for boundary smoothing based on relabeling and vertex relocation.