Three-dimensional image construction method and device and device having storage function

Abstract

The invention discloses a three-dimensional image construction method. The method comprises the following steps: acquiring multiple frames of depth images and displaying each frame of depth image; detecting whether the user selects the point in the displayed depth image and determining the detected and selected depth image as a key frame image and determining the remaining depth images as normal frame images; extracting the plane in each frame of depth image and acquiring the feature information of the extracted plane; registering at least a part of adjacent frames of depth images of multipleframes of depth images by using the extracted feature information of the plane; and constructing the three-dimensional image through fusion of the registration result. With application of the method,the main plane structure can be extracted stably and the more accurate indoor three-dimensional image can be constructed. The invention also provides a three-dimensional image construction device anda device having the storage function.

Description

technical field [0001] The present application relates to the field of image construction, in particular to a method, equipment and storage function for three-dimensional image construction. Background technique [0002] 3D image construction and reconstruction is a challenging research topic, which mainly involves theories and technologies in computer vision, computer graphics, pattern recognition, optimization and other fields. There are many ways to realize 3D image construction. The traditional method is to use laser, radar and other ranging sensors or structured light technology to obtain the structural information of the scene or object surface to construct 3D images. However, most of these instruments are expensive and not easy to carry, so the application Occasions are limited. [0003] In recent years, with the development of computer vision technology, more and more researchers have begun to use pure vision methods to construct 3D images. However, in the prior art...

AI Technical Summary

Problems solved by technology

There are many ways to realize 3D image construction. The traditional method is to use laser, radar and other ranging sensors or structured light technology to obtain the structural information of the scene or object surface to construct 3D images. However, most of these instruments are expensive and not easy to carry, so the application limited occasions

[0003] In recent years, with the development of computer vision technology, more and more researchers have begun to use pure vision methods to construct 3D images. However, in the prior art, most of the matching calculations are performed on all points in the whole image, and the amount of calculation is too large. Large, and due to the lack of plane constraints and optimization, the final reconstruction data can easily lead to wrong data

Moreover, due to the complexity of reconstructing the site, for the simple matching method based on point features, it is easy to lose important features, which ultimately leads to the inability to obtain accurate 3D image data.

Images