A three-dimensional reconstruction method of single-frame encoded structured light based on multiple constraints

Abstract

The invention relates to a three-dimensional reconstruction method of single-frame encoded structured light based on multiple constraints, comprising the following steps: S1 projects single-frame encoded structured light onto a surface of a measured object through a projector, and restricts the single-frame encoded structured light projected by the projector to a horizontal core line; S2, photographing a coded image of a single frame coded structured light deformed by modulating the surface of the measured object by a camera; 3, decoding that encoded image by adopting three constraint relations of column direction constraint, horizontal kernel line constraint and encoding sequence, and obtaining the positions of each encode point in the encoded image; S4, calculating three-dimensional point cloud data of the surface of the measured object according to the pre-calibrated system parameters and combining the positions of each coding point in the coding image. The invention can improve theprecision and the resolution of the three-dimensional reconstruction and reduce the error rate.

Description

technical field [0001] The invention relates to the field of three-dimensional reconstruction, in particular to a multi-constraint-based single-frame coding structured light three-dimensional reconstruction method. Background technique [0002] In the three-dimensional reconstruction technology of active vision, according to the different forms of the projected light source, it is divided into three modes: point, line and surface structured light. Modulation of features. In the two-dimensional surface structured light mode, a projector is often used to project the structured light pattern containing encoded information onto the measured scene or the surface of the target object. At this time, the projector is equivalent to a reverse camera, and there is no need to For multiple scans, you only need to use a camera or other optical sensors to take an image of the measured scene or object, and obtain the three-dimensional surface of the measured object through feature detectio...

AI Technical Summary

Problems solved by technology

Typically, the two-dimensional extended M-arrays of the DeBrujin sequence often combine strips, thin lines, squares, dots, simple geometric shapes designed by ourselves, and other basic elements that are easy to identify for three-dimensional reconstruction. This method only needs to project one frame of grating, It meets the needs of dynamic measurement, but its measurement resolution is relatively low, and there are certain requirements for the continuity of the object surface

Images