Mobile terminal three-dimensional reconstruction and model restoration method and system

Abstract

The invention relates to a mobile terminal three-dimensional reconstruction and model restoration method and system, and belongs to the technical field of image processing and three-dimensional reconstruction. The method comprises the following steps: S1, collecting a series of image data of a modeling object according to a specified angle, and enabling a camera to focus on the modeling object tofuzzify the background in the process; S2, extracting image feature points by using scale invariant feature transform (SIFT), fusing the feature points of every two images, and maximizing the resolution of the images when the feature points are extracted; S3, completing sparse reconstruction by using a motion recovery structure SFM; S4, completing dense reconstruction by using a stereoscopic vision method MVS; during dense reconstruction, the minimum pixel point threshold value is reduced, so that the effective point cloud is increased; S5, finishing point cloud modeling by using a Poisson reconstruction method PSR; during Poisson reconstruction, shear elements are reduced, so that the effective surface is increased; S6, topology reconstruction is carried out; S7, geometric correction; andS8, the model is displayed, and 3D printing or 3D engraving is conducted.

Description

technical field [0001] The invention belongs to the technical field of image processing and three-dimensional reconstruction, and relates to a mobile terminal three-dimensional reconstruction and model restoration method and system. Background technique [0002] In recent years, mobile 3D printing and 3D engraving technologies have developed rapidly, and various 3D modeling and 3D printing apps emerge in endlessly. There are two main types of mobile 3D reconstruction technologies: one is through feature matching and estimation of camera and scene geometry, for example, through motion recovery structure (structure from motion, SFM) or color matching (for example, dense tracking and mapping (dense tracking and mapping, DTAM)), these techniques are sensitive to matching, and are ineffective in homogeneous areas, and the reconstructed model has many defects, which cannot be directly used for 3D printing or 3D engraving. The other is to use the depth and texture information of t...

AI Technical Summary

Problems solved by technology

The existing 3D reconstruction method combined with SFM and MVS also has the following defects: First, SFM is an algorithm for feature matching and estimation of camera and scene geometry. This algorithm is sensitive to matching and is invalid in homogeneous areas. Modeling is successful The rate and modeling defect rate are completely dependent on the quality of the input image data. However, due to the randomness of shooting objects and the diversity of their cameras in real life, the quality of image data cannot be guaranteed.

Secondly, due to the inherent defects of the algorithm, even if the quality of the image data meets the requirements, the final modeling result must be repaired later in order to output a defect-free model with no holes, no patches, and a complete closure.

Images