Visual positioning method and system

Abstract

The embodiment of the invention provides a visual positioning method and system. The visual positioning method comprises the steps of obtaining a real-time binocular image of a local scene and a binocular image set of a previous time window of the real-time binocular image; constructing a local visual point cloud picture according to the real-time binocular image and the binocular image set; and matching the local visual point cloud picture with the global visual point cloud picture, and obtaining camera position information and camera attitude information according to a matching result. According to the embodiment of the invention, the real-time binocular image is acquired through the binocular vision camera; a local visual point cloud picture is constructed according to the real-time binocular image and the binocular image in the previous time sequence. According to the embodiment of the invention, deployment and operation are easier, the visual positioning efficiency is improved, meanwhile, more abundant visual texture images and visual geometric information are provided, and the application range of visual positioning is expanded.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of visual positioning, and in particular, to a visual positioning method and system. Background technique [0002] In the existing visual positioning technology, the laser is used as a sensor to obtain 3D point cloud data, and the positioning in the known 3D point cloud map is realized. [0003] However, there are many defects in the laser sensor itself, which limit the application of the visual positioning system. For example, the laser sensor is generally large in size, which makes it impossible to install and deploy in many application scenarios, which limits the large-scale application of the visual positioning system. . In addition, the laser sensor itself acquires spatial point cloud data. The information contained in these data is not rich enough, and it is difficult to use it for other purposes except for positioning, so more function expansion and development cannot be carried...

AI Technical Summary

Problems solved by technology

[0003] However, there are many defects in the laser sensor itself, which limit the application of the visual positioning system. For example, the laser sensor is generally large in size, which makes it impossible to install and deploy in many application scenarios, which limits the large-scale application of the visual positioning system.

In addition, the laser sensor itself acquires spatial point cloud data, the information contained in these data is not rich enough, and it is difficult to use it for other purposes except for positioning, so more function expansion and development cannot be carried out

Images