Indoor scene three-dimensional point cloud reconstruction method and system based on depth information fusion

Abstract

The invention discloses an indoor scene three-dimensional point cloud reconstruction method based on depth information fusion. The method comprises the following steps: collecting a color image and a depth image of a scene; fusing the color image and the depth image to perform feature extraction and matching; determining the pose of the depth camera and screening out a key frame; generating a three-dimensional point cloud of each key frame according to the key frames; and splicing the three-dimensional point cloud of each key frame frame by frame according to the pose of the depth camera to obtain a globally consistent scene three-dimensional point cloud model. According to the method, whether a point is a feature point is judged by using image brightness information and depth information at the same time in feature extraction, so that the extracted feature points have high representativeness. Meanwhile, homogenization operation is performed on the extracted feature points, it is ensured that the number of the feature points is within a reasonable range, the calculated amount of a system is effectively controlled. The system reliability is improved. The accuracy of the front end is improved, the robustness of the front end is enhanced, and the result of back-end optimization is more accurate or the number of iterations of back-end optimization is reduced.

Description

technical field [0001] The present invention relates to the technical fields of computer graphics, computer vision, and robotics, and more specifically, to a method and system for reconstructing three-dimensional point clouds of indoor scenes based on depth information fusion. Background technique [0002] In recent years, the problem of simultaneous localization and mapping (SLAM) has once again become a hot topic in the field of computer vision and robotics research. SLAM can be divided into two parts for discussion, that is, positioning and mapping. Positioning refers to the description of the robot's own position and posture, while mapping is the description of the surrounding environment information of the robot. When the robot knows its own position and the surrounding environment Once information is available, robots can help humans complete many tasks. With the rapid development of computer vision technology, the cost of expensive products such as cameras and laser ...

AI Technical Summary

Problems solved by technology

At present, the traditional visual odometry uses feature point methods such as ORB, SIFT, SURF or their variants. In the process of feature extraction, characteristics such as illumination, rotation, and scale are considered, but these methods do not take into account the important feature of pixel depth value.

Images