A depth information acquisition system based on a dual-structure light RGB-D camera

Abstract

The invention relates to a depth information acquisition system of a dual-structure light RGB-D camera. The system comprises dual-structure light RGB-D camera structure comprising two structured lights RGB-D sub-cameras symmetrically arranged at 180 degrees, a RGB-D camera calibration module which is used for realizing structured light RGB-D camera internal parameters and distortion parameters external parameters between the color camera and the depth camera according to the asymmetric calibration plate; a multi-source depth information acquisition module which is used for acquiring two original depth images of the two sub-cameras in an asynchronous manner, and forming a binocular image by using the color images acquired by the two cameras to generate a third depth image; and a multi-source depth information processing module which is used for realizing the alignment of the three depth images and the color images, fusing the three depth images, and taking the aligned color images and the fused depth image as the output of a final system. Compared with the prior art, the method has the advantages of parameter calibration, solving of a depth information missing hole, improvement of the quality of depth data, high real-time performance and the like.

Description

technical field [0001] The invention relates to the field of 3D vision, in particular to a depth information acquisition system based on a dual structured light RGB-D camera. Background technique [0002] In recent years, domestic research on 3D vision has become more and more popular. 3D cameras can obtain the distance from the camera to every point on the surface of the object, perceive the shape and outline of the object, and obtain richer information. The quality of depth information in RGB-D information has an important impact on the entire 3D vision system. [0003] The current mainstream 3D cameras can be divided into two types: one is the 3D vision sensor with high price and good depth data quality represented by laser sensor, such as SICK-about 300,000, Enshape-about 300,000, Ensenso-about 100,000, Their common feature is high precision, but slow and expensive, which is not suitable for large-scale promotion in the industrial field; the other is a series of consume...

AI Technical Summary

Problems solved by technology

[0006] To sum up, the existing methods of obtaining depth images are either expensive and have no potential for general application; or the quality of the obtained depth images is poor enough to meet the needs of some industrial applications, both of which limit 3D The wide application of vision is not conducive to the construction of 3D vision system

Images