Diamond Inclusion Reconstruction
- PDF / 2,225,425 Bytes
- 19 Pages / 439.37 x 666.142 pts Page_size
- 75 Downloads / 146 Views
Diamond Inclusion Reconstruction Zijian Yuan1 · Yanzhi Song1 · Falai Chen1 · Zhouwang Yang1 Received: 25 February 2019 / Revised: 24 September 2019 / Accepted: 19 November 2019 © School of Mathematical Sciences, University of Science and Technology of China and Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract In this paper, we make an improvement on the conventional visual hull reconstruction method that runs on a single consumer graphics card. The target application is a high-precision diamond inclusion reconstruction system. One major contribution of this paper is an evaluation system of voxels for high-precision reconstruction. In contrast to existing approaches, it allows us to reconstruct the thin structure of the diamond inclusion. Based on this, we obtain a significant improvement in reconstruction precision, especially for thin inclusion. Keywords 3D reconstruction · Voxel model · CUDA Mathematics Subject Classification 65D18 · 94A08
1 Introduction In this paper, we focus on the reconstruction of rough diamond inclusion. In the problem of cutting rough diamond, we want to remove the inclusion of the rough diamond to get the diamonds without inclusion. Besides, in order to maximize the economic value of the rough diamond, we are required to plan the retained part as large as possible while keeping this part of rough diamond away from the inclusion. In order to achieve the two goals, we need to reconstruct the structure of the diamond
B
Zhouwang Yang [email protected] Zijian Yuan [email protected] Yanzhi Song [email protected] Falai Chen [email protected]
1
School of Mathematical Sciences, University of Science and Technology of China, Hefei 230026, People’s Republic of China
123
Z. Yuan et al.
inclusion and the surface of the rough diamond. Since the clarity is an important indicator for assessing the value of diamonds, this work requires a highly precise reconstruction model of the inclusion. In this problem, we are required to reconstruct the model of the diamond inclusion based on the images of the diamond. This can be considered as a multi-view reconstruction problem. There are already a number of high-quality algorithms focused on the multi-view reconstruction problem. Many algorithms in this field have already been able to reconstruct both the shape and the texture consistent with the views [2,4]. But all these algorithms highly rely on the feature points or color details in images to update the model by iteration. In our reconstruction problem, due to the limitation of the acquisition of the rough diamond images, there is little such detail in the images. The difficulty in acquisition of feature points from the images limits the application of these algorithms to our problem. However, the visual hull method is able to reconstruct the model from the images with little feature information. Visual hull model was proposed by Matusik et al. [6] firstly. Instead of computing an explicit 3D model, for each view from the camera, the visual hull method emanates viewi
Data Loading...
