Super-high-resolution three dimensional reconstruction method based on global variation technology

Abstract

The invention discloses a super-high-resolution three dimensional reconstruction method based on a global variation technology. The method comprises the following steps: low-resolution three dimensional geometrical model and a high-resolution surface normal graph of a scene to be constructed is obtained; a global optimization energy function is constructed and obtained according to the difference between the low-resolution geometrical model and the high-resolution normal graph; the minimum value of the global optimization energy function is solved, a depth value of each pixel point is obtained, and three dimensional reconstruction is carried out according to the depth values of all the pixel points. The super-high-resolution three dimensional reconstruction method based on non-local total variation regular constraint condition lowers dependence of a final reconstruction result on an original result effectively and raises the three dimensional reconstruction precision greatly.

Description

technical field [0001] The invention relates to the field of computer vision three-dimensional reconstruction, in particular to an ultra-high resolution three-dimensional reconstruction method. Background technique [0002] Three-dimensional scanning technology is a measurement method that obtains the three-dimensional shape and position data of the measured object through certain detection means. The current three-dimensional scanning technology can be roughly divided into two categories: contact and non-contact. The main products of contact 3D scanning are mechanical three-coordinate measuring machines, which have high measurement accuracy, but the measurement speed is slow, and only one point can be measured at a time, so it is difficult to achieve online detection, and the measurement object cannot be soft or easy to scratch. injured object. [0003] Optical scanning measurement is the use of optical principles, visual scanning technology and image processing methods t...

AI Technical Summary

Problems solved by technology

[0005] Surface structured light 3D scanning can obtain high-precision 3D data on the surface of objects, but its effect on the acquisition of 3D data of surface fine features is not very satisfactory, because surface structured light 3D scanning requires hardware devices such as cameras and projectors, and requires To obtain accurate three-dimensional data of the subtle features of objects, it is necessary to increase the collection density of three-dimensional data, which can generally be achieved by increasing the resolution of the image. Although high-resolution cameras are easy to obtain, high-resolution projectors are not. Easy to find, which restricts the improvement of the accuracy of surface structured light 3D scanning

[0006] The 3D reconstruction based on photometric stereo vision can obtain the detailed features of the object surface very well, but the data it obtains is only 3D data in units of pixels, and when the image quality is not good or the illumination model assumption is not appropriate, it can be obtained by The 3D data reconstructed by photometric stereo vision still has a large deformation

In this way, to obtain high-density data is subject to certain restrictions

Images