Indoor space panoramic image generation method based on multi-sensor fusion

Abstract

The present invention provides a multi-sensor fusion-based indoor space panoramic image generation method capable of generating a panoramic image with depth information and clear image resolution, the method includes: moving the survey vehicle based on a predetermined movement rule to obtain an image of each station or point cloud data collection; merge the images collected by each image collector with point cloud data to generate corresponding high-quality depth maps; construct a virtual image center, and use the depth value of each depth map as the projection distance for spherical Projection to obtain the corresponding registration image; fill holes in the bottom of the image of the current station according to the image of the previous station and the image of the next station; stitch and fuse the overlapping areas of the registration image, and finally output the image with depth information and resolution High-efficiency and clear panoramic images, which can realize functions such as measurement, 3D reconstruction, and visual positioning.

Description

Technical field [0001] The present invention belongs to the field of image mapping processing technology, specifically relates to a panoramic image capable of generating a panoramic image with depth information and a clear image resolution based on multi-sensor fusion indoor space panoramic image generation method. Background [0002] Digital twins make full use of physical models, sensor updates, operating history and other data, integrate multi-disciplinary, multi-physical quantity, multi-scale, multi-probability simulation process, and complete the mapping in the virtual space, thereby reflecting the whole life cycle process of the corresponding physical equipment. The panorama is the key digital data of the indoor digital twin system; the panorama, with rich color information, and has a strong sense of immersion for the viewer, which plays a key role in the monitoring, digital city, and medical image analysis of large public places such as airports or railway stations. [00...

AI Technical Summary

Problems solved by technology

[0003] However, most of the panoramas currently on the market can only be browsed without depth information, that is, scale information, and operations such as measurement and 3D reconstruction cannot be performed through panoramic images; moreover, due to the large indoor parallax, when multi-camera platforms perform panorama stitching, if Without accurate depth information, mis-layers and ghosting are prone to occur at the stitching, resulting in low quality panoramic images

Images