Visual odometer pose optimization method based on highlight pixel detection

A technology of visual odometry and optimization method, applied in the directions of calculation, image enhancement, image analysis, etc., can solve the problems of image matching error and decreased positioning accuracy, and achieve the effect of improving positioning accuracy and avoiding high-light pixels.

Active Publication Date: 2021-07-09
ZHEJIANG LAB
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the position of the highlight pixel will change with the movement of the camera angle of view and the position of the light source, resulting in the matching error of adjacent images in visual positioning, resulting in a decrease in positioning accuracy

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
  • Visual odometer pose optimization method based on highlight pixel detection
  • Visual odometer pose optimization method based on highlight pixel detection
  • Visual odometer pose optimization method based on highlight pixel detection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0043] Specific embodiments of the present invention will be described in detail below in conjunction with the accompanying drawings. It should be understood that the specific embodiments described here are only used to illustrate and explain the present invention, and are not intended to limit the present invention.

[0044] Such as figure 1 Shown, concrete implementation of the present invention is as follows:

[0045] S1: The camera is mounted on a mobile platform, its position and angle of view are fixed, and the environmental color image information is obtained in real time according to a fixed number of frames;

[0046] S2: Calculate the highlight pixel image in the color image: Calculate the minimum intensity of the color image I min , the maximum intensity I max , chromaticity value Λ psf ( i ), introduce the corresponding chromaticity values ​​of all pixels into the minimum chromaticity-maximum chromaticity two-dimensional space for clustering, and calculate a...

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 visual odometer pose optimization method based on highlight pixel detection. The method mainly comprises the steps of highlight pixel detection, weight matrix calculation, feature descriptor calculation and least square optimization. Mirror reflection often occurs on the surface of a metal and partially smooth object, resulting in the presence of highlight pixels in an image captured by a camera, and the highlight pixel position can change along with the movement of the visual angle of the camera, so that the adjacent images are wrongly matched in visual positioning, and the positioning precision is reduced. The method is characterized in that highlight pixel detection is introduced, a detection result is converted into a weight matrix to be added into the in-place pose optimization process, wrong matching caused by highlight pixels is eliminated, and the precision of visual positioning is effectively improved.

Description

technical field [0001] The invention relates to the field of robot vision positioning, in particular to a method for optimizing the pose of a visual odometer based on highlight pixel detection. Background technique [0002] Specular reflections often occur on the surface of smooth objects such as metals, resulting in high-light pixels in images captured by the camera. However, the position of the highlight pixel will change with the movement of the camera angle of view and the position of the light source, resulting in the matching error of adjacent images in visual positioning, resulting in a decrease in positioning accuracy. Contents of the invention [0003] In order to solve the deficiencies in the prior art, the present invention introduces a highlight weight matrix in the traditional visual odometry pose solution to achieve the purpose of improving the positioning accuracy of the visual odometry. The present invention adopts the following technical solutions: [000...

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): G06T7/73
CPCG06T2207/10016G06T2207/10024G06T7/73
Inventor 宋伟王程朱世强廖建峰郑涛
Owner ZHEJIANG LAB
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