An image-based diamond raw stone inclusion reconstruction method

Abstract

The invention discloses an image-based diamond raw stone inclusion reconstruction method. The method comprises the following steps: positioning a rotating shaft; cutting the picture; pre-processing the picture to reduce the interference of the background of the picture on a reconstruction result; estimating the light transmission of voxels; model extraction: extracting a voxel model of the diamondraw stone inclusions and a surface model of the diamond raw stone based on the light transmission of voxels; calculating an approximation model; Model output. The invention provides an image-based diamond raw stone inclusion reconstruction method. The method comprises the following steps: preprocessing an input diamond raw stone inclusion image; evaluating the light transmission of voxels in a three-dimensional space by using the gray value of the processed image; extracting a voxel model as a three-dimensional model of inclusions based on the light transmission degree; and then the three-dimensional model is subjected to approximate calculation, so that the reconstruction task of the diamond raw stone inclusions is completed, the reconstruction precision of the diamond raw stone inclusions is improved, and a solid foundation is laid for improving the utilization efficiency of the diamond raw stone in the diamond raw stone cutting planning and shortening the required time for designing a cutting scheme.

Description

technical field [0001] The invention belongs to the technical field of three-dimensional reconstruction, in particular to an image-based reconstruction method for inclusions in rough diamonds. Background technique [0002] Multi-view image 3D model reconstruction is a process of reconstructing a 3D model of an object based on object images taken from multiple angles. [0003] Viewing angle changes, occlusion, and background interference of the image content are all difficulties in the reconstruction task. In addition, the structure of the inclusions in rough diamonds is complex, and some inclusions are very small and difficult to identify. Therefore, the reconstruction of rough diamond inclusions is a difficult task. The problem. The traditional multi-view reconstruction method generally extracts the feature points in the image first, and then performs modeling based on feature point matching. Due to the chaotic structure of the inclusions in rough diamonds, the extraction...

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide an image-based method for reconstructing the inclusions of rough diamonds. By preprocessing the images of inclusions in rough diamonds taken from multiple perspectives, the gray value of the processed images is used to evaluate the transparency of voxels in three-dimensional space. Light, and based on the degree of light transmission, extract a voxel model as the three-dimensional model of the inclusions, and then perform approximate calculations on the three-dimensional model to achieve the task of reconstructing the inclusions of rough diamonds, which solves the problem of existing methods for reconstructing models The problem of low quality reconstruction model

Images