Detection method of natural target in robot vision navigation

A technology of robot vision and target detection, which is applied in the field of natural target detection of robot visual navigation, can solve the problems that it is difficult to describe and distinguish different targets, and it is difficult to achieve natural target detection of robot visual navigation, so as to achieve fast calculation speed, good practical value, The effect of improving robustness

Inactive Publication Date: 2006-12-06
INST OF AUTOMATION CHINESE ACAD OF SCI
View PDF0 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

When the target can only partially appear in the image, or there are strong interference targets in the image, it is difficult to describe and distinguish different targets
[0007] To sum up, the existing target detection methods are still difficult to meet the following requirements of natural target detection in robot vision navigation:

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
  • Detection method of natural target in robot vision navigation
  • Detection method of natural target in robot vision navigation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] Specific implementation plan

[0034] The block diagram of the natural target detection method in the robot visual navigation is as follows figure 1 Shown. The target detection process consists of eight steps: target modeling, image acquisition, preprocessing, segmentation parameter selection, natural image segmentation, feature extraction, target authentication, and output results.

[0035] First, establish the target model. The target model adopts statistical modeling. The color models of the target to be detected and the interference target are described by three color components of Cb, Cr and R / G, and the three-dimensional color components are projected to Cb-Cr and Cb-R / G respectively. On the plane, the target color template plane is formed. Through offline learning of a large number of target samples, the samples are mapped to the target color template plane, and the color templates of the target to be detected and the interference target are obtained by statistics. ...

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

This invention relates to the goal monitor technology field, specially a kind of natural goal examination method used in the robot vision navigation. The method includes: goal modeling, picture gathering, pretreatment, division parameter choice, robustness natural picture division, characteristic withdraw, goal authentication, and output result.

Description

Technical field [0001] The invention relates to the technical field of target detection, in particular to a natural target detection method for robot visual navigation. Background technique [0002] The detection of natural targets is a key problem to be solved in vision-based mobile robot navigation. The influence of complex background and irregular illumination changes makes it difficult to extract natural targets from the background. However, compared with artificial targets, visual navigation based on natural target detection has broader application prospects. [0003] Object detection can be broadly divided into methods based on touch, hearing, taste and vision. In robot vision navigation, target detection is based on vision-based image target detection. Image target detection can be further divided into moving target detection and static target detection. The present invention is directed to static natural target detection for navigation. [0004] Static natural target dete...

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): G06K9/32
Inventor 易建强洪义平赵冬斌
Owner INST OF AUTOMATION 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