一种基于空间编码结构光的稠密三维重建算法
首发时间:2020-04-01
摘要:在本文中,基于空间编码结构光提出了一种新颖、简单、快速的稠密三维重建方法。首先,针对彩色条纹不适用于带颜色物体的问题,提出了一种由正弦条纹和伪随机点组合成的结构光编码模式,所有的条纹均采用相同的颜色进行编码,通过引入伪随机点来解决条纹的识别问题;其次,针对传统条纹方法获得的点云空间分辨率低的问题,提出了一种基于局部相位匹配的稠密三维重建方法,基于条纹中心点提取算法,准确提取条纹的中心坐标,利用傅立叶变换获取条纹间的相位信息,最后对相邻条纹间的所有像素点进行匹配,实现稠密三维重建。实验结果表明,在被测物体表面结构复杂的条件下,本文提出的算法依然能够有效地获取物体表面的三维信息,且具有较好的精度和较高的空间分辨率。
For information in English, please click here
A Dense Three-dimensional Reconstruction Algorithm Based on Spatially Encoded Structured Light
Abstract:In this paper, we propose a novel, simple and fast dense three-dimensional reconstruction method based on spatially encoded structured light. Firstly, for the problem that color stripe-based mothed are not suitable for colored objects, a structured light coding pattern composed of sinusoidal stripes and pseudo-random dots is proposed. All the stripes are coded with the same color, and pseudo-random dots are introduced to identify the stripes. Secondly, aiming at the low spatial resolution of the point cloud obtained by the traditional stripe-based mothed, a dense three-dimensional reconstruction method based on local phase matching is proposed. The center points extraction algorithm is used to accurately extract the stripes. Obtain the phase of pixels between adjacent stripes by Fourier transform. Finally, all the pixels between adjacent stripes are matched to achieve dense three-dimensional reconstruction. The experimental results show that the proposed algorithm in this paper can still effecA Dense Three-dimensional Reconstruction Algorithm Based on Spatially Encoded Structured Lighttively acquire the three-dimensional information of the object surface under the condition that the surface structure of the measured object is complex, and has better accuracy and higher spatial resolution.
Keywords: 3D reconstruction Structured light Projector-cameras system Fourier transform
引用
No.****
动态公开评议
共计0人参与
勘误表
一种基于空间编码结构光的稠密三维重建算法
评论
全部评论0/1000