Three-dimensional object recognition system and method based on laser radar and monocular vision

A laser radar and 3D target technology, applied in the field of computer vision, can solve problems such as difficulty in 3D point cloud target recognition, and achieve wide applicability, robustness, and high recognition rate

Active Publication Date: 2018-12-04
XI'AN INST OF OPTICS & FINE MECHANICS - CHINESE ACAD OF SCI
View PDF5 Cites 43 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] The present invention provides a three-dimensional target recognition system and method based on laser radar and monocular vision, aiming to solve the problem of difficult target recognition in existing three-dimensional point clouds

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
  • Three-dimensional object recognition system and method based on laser radar and monocular vision
  • Three-dimensional object recognition system and method based on laser radar and monocular vision
  • Three-dimensional object recognition system and method based on laser radar and monocular vision

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] This embodiment is a 3D target recognition system applied to a robot vision system, which acquires a scene image from a monocular camera, recognizes the target and inversely calculates its spatial position in the 3D point cloud acquired by the laser radar, thereby providing the scene for the robot The spatial location information of the target.

[0046] Such as figure 1 As shown, the 3D target recognition system mainly includes the following six modules:

[0047] 1. Point cloud acquisition module. This module is responsible for acquiring the 3D point cloud of the scene from LiDAR.

[0048] 2. Image acquisition module. This module is responsible for getting the scene image from the camera.

[0049] 3. Point cloud projection module. This module is responsible for projecting the 3D point cloud of the scene acquired by the lidar. The purpose of the projection is to obtain a 2D image that can be directly matched with the subsequent modules.

[0050] 4. Image matching m...

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 invention discloses a three-dimensional object recognition system and method based on laser radar and monocular vision. The method comprises the following steps: projecting point cloud onto a two-dimensional image, extracting and matching features of the projected image and a monocular image, and calculating a projective transformation relationship between the projected image and the monocularimage; carrying out object recognition through rich color information of the monocular image and extracting an object area of interest; inversely calculating corresponding point cloud blocks of the object in laser radar point cloud through a transformation relationship between laser radar and a monocular, and finally outputting space position information of the point cloud blocks. The three-dimensional object recognition system and method overcome the disadvantages of difficult recognition of a three-dimensional point cloud object, and achieve recognition and positioning of an object in a three-dimensional scene by means of the mature technique of two-dimensional image object recognition and the high-precision ranging performance of laser radar.

Description

technical field [0001] The invention belongs to the technical field of computer vision, and relates to a three-dimensional target recognition system and method based on a laser radar point cloud. Background technique [0002] As lidar technology is used in more and more fields, 3D target recognition based on lidar point cloud is also receiving more and more attention. For example, in the fields of missile guidance, satellite tracking, and robot navigation, it is necessary to use lidar to scan, identify targets and output the spatial position information of the target. Specifically, lidar is used as a sensor to identify and extract the acquired 3D point cloud target. Since the 3D point cloud acquired by lidar is disordered, noisy, and contains many missing points, it has always been a difficult problem to directly perform target recognition on the 3D point cloud. [0003] At present, the methods for 3D point cloud target extraction mainly include: target extraction method ba...

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 Applications(China)
IPC IPC(8): G06K9/00
CPCG06V20/647G06V20/13
Inventor 曹剑中黄会敏周祚峰郭惠楠胡国良
Owner XI'AN INST OF OPTICS & FINE MECHANICS - CHINESE ACAD OF SCI
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

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap