Monocular real-time target recognition and pose measurement method

A target recognition, pose measurement technology, applied in the field of computer vision, can solve the problem of high system complexity, achieve the effect of improving accuracy, reducing algorithm and system complexity

Inactive Publication Date: 2015-03-25
SHANDONG UNIV
View PDF4 Cites 45 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This type of method needs to set characteristic ta

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
  • Monocular real-time target recognition and pose measurement method
  • Monocular real-time target recognition and pose measurement method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] The present invention is described in detail below in conjunction with accompanying drawing:

[0026] Such as figure 1 As shown, the present invention first utilizes the improved ORB algorithm to obtain the feature point data of the target object, and stores them in the target database together with the three-dimensional world coordinates and image coordinates of the four vertices of the target plane; then the camera is calibrated to obtain the camera model parameters; For the recognized target, the camera extracts the ORB feature points of the real-time scene, matches the feature points of the target image in the database with the feature points of the real-time scene, and removes the mismatch through random sampling consistency detection to obtain a stable homography matrix (Homography). The image coordinates of the four vertices of the target plane in the matrix and the database are obtained to obtain the image coordinates of the four vertices of the target in the sc...

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 monocular real-time target recognition and pose measurement method. The method comprises the steps that various target object feature point data are obtained, and the target object feature point data and three-dimensional world coordinates and image coordinates of the four vertexes of a target plane are stored into a target database; a target to be recognized is selected, and target feature point data and target information are extracted from the target database; a video camera extracts real-time scenes, image preprocessing is carried out on the extracted real-time scenes, real-time scene feature points are extracted, and real-time image feature point data are obtained; target image feature points in the target database and the real-time scene feature points are matched; a homography is obtained through calculation through the matched feature point data, and target recognition is completed; pose measurement and position determination of the target are completed. According to the monocular real-time target recognition and pose measurement method, the speed advantage of an ORB algorithm is utilized, the single video camera is used as the sensing means, and real-time recognition and pose measurement of the moving target are achieved.

Description

technical field [0001] The invention relates to the field of computer vision, in particular to a monocular real-time target recognition, attitude measurement and position determination method. Background technique [0002] In order to realize the perception of objects, intelligent robots must have the ability to recognize objects and have the ability to recognize the posture of objects. It is a basic function of intelligent robots to recognize target objects from complex scenes, and to select the appropriate grasping method according to the posture of the object, just like humans grasp objects. Therefore, designing a visual positioning system, which can not only accurately identify the target object, but also judge its posture, has a very broad application prospect. [0003] The key to the research of vision-based target recognition is the feature representation method of the target. The representation method must not only uniquely describe a target, but also have commonali...

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
IPC IPC(8): G06K9/00G06K9/46
CPCG06V20/20
Inventor 杨明强崔振兴曾威陆中州
Owner SHANDONG UNIV
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