Registration method, device, equipment and medium of laser point cloud and 2D image

A laser point cloud and image technology, applied in the field of intelligent transportation, can solve problems such as the inability to find points with the same name, the lack of significant improvement in the registration accuracy of laser point clouds and high-precision images, and matching errors

Active Publication Date: 2021-06-04
BEIJING BAIDU NETCOM SCI & TECH CO LTD
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] Since the laser point cloud and the high-precision image contain information of different attributes, there is still a matching error between the same-named points (that is, the same target point in different data sources) matched by the above-mentioned image registration method, and even the same-named point cannot be found. As a result, the registration accuracy of laser point clouds and high-precision images has not been significantly improved

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Registration method, device, equipment and medium of laser point cloud and 2D image
  • Registration method, device, equipment and medium of laser point cloud and 2D image
  • Registration method, device, equipment and medium of laser point cloud and 2D image

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0035] figure 1 It is a flow chart of the registration method of laser point cloud and 2D image provided by the first embodiment of the present invention. In the case of 2D image registration, the method can be implemented by a device for registering laser point clouds and 2D images, which can be implemented in software and / or hardware, and can be integrated in any computing device, including but not Limited to servers, on-board computing devices, etc.

[0036] During the registration process of laser point cloud and 2D image, the laser point cloud data and 2D image can be continuously received in the form of data stream or image stream. Taking the laser point cloud data and 2D image obtained through data collection as an example, the registration process of the laser point cloud and 2D image will be described in detail. Each semantic image mentioned in the embodiment refers to the semantic image of the laser point cloud and the semantic image of the 2D image that participat...

Embodiment 2

[0053] image 3 It is a flow chart of the registration method of laser point cloud and 2D image provided by Embodiment 2 of the present invention. This embodiment further optimizes and expands on the basis of the above embodiment. Such as image 3 As shown, the method may include:

[0054] S210, for the 2D semantic image of the laser point cloud and the semantic image of the 2D image, respectively determine the target area including the lane line on each semantic image, and determine the disappearance on each semantic image according to the direction of the lane line in the target area point.

[0055] In this embodiment, the vanishing point refers to a virtual intersection point of multiple fitted straight lines after the near and far straight lines are respectively fitted to the multiple lane lines on the semantic image. The size of the target area can be flexibly set under the condition that the overall extension trend of each lane line from near to far on each semantic i...

Embodiment 3

[0077] Figure 4 It is a flow chart of the registration method of laser point cloud and 2D image provided by Embodiment 3 of the present invention. This embodiment is further optimized and expanded on the basis of the above embodiments. Such as Figure 4 As shown, the method may include:

[0078] S310. Perform pixel classification on the laser point cloud image in 3D space to obtain a 3D semantic image of the laser point cloud, and use the initial projection relationship between the laser point cloud and the 2D image to perform projection transformation on the 3D semantic image to obtain the 2D semantic image of the laser point cloud. image.

[0079] Among them, the laser point cloud data includes the 3D spatial information of each object point in the driving environment. After pixel classification based on the intensity information of the laser point cloud in the 3D space, the 3D semantic image of the laser point cloud can be obtained, and then projected to 2D Space, get t...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The embodiment of the present invention discloses a registration method, device, device and medium for laser point cloud and 2D image, wherein the method includes: on the 2D semantic image of the laser point cloud and the semantic image of the 2D image, respectively According to the intersection of each lane line in each set of lane line pairs and the preset graphics on the corresponding semantic image, determine the same name on the laser point cloud and 2D image Point pair; use the point pair with the same name to determine the registration correction matrix between the laser point cloud and the 2D image; use the initial projection relationship between the laser point cloud and the 2D image, and the registration correction matrix to register the laser point cloud and the 2D image. The embodiment of the present invention can solve the problem of low registration accuracy between laser point cloud and high-precision image in the prior art, can realize accurate registration between laser point cloud and 2D image, and ensure the fusion accuracy of the two.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of intelligent transportation, and in particular to a registration method, device, equipment and medium of a laser point cloud and a 2D image. Background technique [0002] Image registration (Image Registration) technology is the process of matching and superimposing two or more images acquired at different times, different sensors (imaging devices) or under different conditions (illuminance, camera position and angle, etc.), and is currently widely used. It is used in remote sensing data analysis, computer vision and image processing and other fields. [0003] In the process of making high-precision maps, the laser point cloud acquisition equipment is used to obtain the spatial position information of the target, and the high-definition color camera is used to obtain the color information of the target. Laser point cloud acquisition equipment and high-definition color cameras have th...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Patent Type & Authority Patents(China)
IPC IPC(8): G06T7/33G06T7/11
CPCG06T2207/10004G06T2207/10028G06T7/11G06T7/337
Inventor 赵祖轩
Owner BEIJING BAIDU NETCOM SCI & TECH CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products