3D point cloud densification method and device based on binocular RGB image

Abstract

The embodiment of the invention provides a 3D point cloud densification method and device based on a binocular RGB image. The method comprises the following steps: step 1, generating a depth image from a binocular RGB image; estimating an approximate three-dimensional coordinate point of each pixel point of the depth image under a LiDAR coordinate system according to the depth information of the depth image; step 2, performing ground segmentation on the point cloud by using a cyclic RANSAC algorithm, and extracting a non-ground point cloud; step 3, inserting the extracted non-ground point cloud into a KDTree, searching a predetermined number of neighbor points in the KDTree according to the approximate three-dimensional coordinate points of each pixel point under the LiDAR coordinate system, and performing curved surface reconstruction by utilizing the neighbor points; and step 4, exporting accurate coordinate points of the approximate three-dimensional coordinate points by a computational geometry method according to a curved surface reconstruction result and calibration parameters of the LiDAR and the camera, and fusing the accurate coordinate points with the original LiDAR pointcloud to obtain a densified point cloud.

Description

technical field [0001] The present invention relates to the field of image processing, in particular to a 3D point cloud densification method and device based on binocular RGB images. Background technique [0002] With the continuous development of technology, research in the field of driverless vehicles has also made great progress. In the field of unmanned vehicles, the detection and recognition of objects on the road ahead is one of its important tasks. LiDAR (Light Detection And Ranging) and cameras, as the two major sensors supporting object detection tasks, are essential in the configuration of unmanned vehicles. [0003] LiDAR provides accurate three-dimensional distance information, so it is widely used. However, the cost of LiDAR increases as the number of emission wires increases, so the number of laser wires of LiDAR equipped on a vehicle is usually small, and the price of a Velodyne HDL-64E is about 4 times that of an ordinary family car, which makes self-drivi...

AI Technical Summary

Problems solved by technology

[0006] That is to say, lidar plays an important role in autonomous driving. Even though it is expensive, the number of laser beams equipped is still small, resulting in sparse point cloud density.

As another sensor that is widely used for object detection, the camera, although the price is low, usually cannot complete the object detection task in three-dimensional space well due to the lack of distance information in two-dimensional image data.

Images