Binocular vision odometer based on direct method and linear feature and measurement method

A measurement method, binocular vision technology, applied in the direction of measuring device, measuring distance, surveying and navigation, etc., can solve the problem of not being able to track well, and achieve the problem of scale uncertainty, good robustness, and not easy to lose Effect

Pending Publication Date: 2019-03-08
轻客小觅机器人科技(成都)有限公司
View PDF7 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] To this end, the embodiment of the present invention provides a binocular visual odometer and measurement method based o

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
  • Binocular vision odometer based on direct method and linear feature and measurement method
  • Binocular vision odometer based on direct method and linear feature and measurement method
  • Binocular vision odometer based on direct method and linear feature and measurement method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0029] This embodiment provides a binocular visual odometer measurement method based on direct method and line feature. Such as figure 1 As shown, the measurement method includes the following steps.

[0030] Step 11. The binocular camera collects images simultaneously to obtain a binocular image.

[0031] The left and right eyes of the binocular camera of the binocular visual odometer simultaneously collect images in real time. The left and right binoculars of the binocular camera can be cameras set left and right.

[0032] Step 12, extracting the image information of the region where the pixel intensity gradient in the binocular image meets a certain threshold, and obtaining the binocular image information;

[0033] In an example, the threshold includes a first threshold; extracting image information of an area where the pixel intensity gradient in the binocular image meets a certain threshold, and obtaining the binocular image information includes: extracting the pixel i...

Embodiment 2

[0051] This embodiment provides a binocular visual odometer based on direct methods and line features. Such as figure 2 As shown, the binocular visual odometer includes:

[0052] The acquisition unit 21 is used for simultaneously acquiring images by binocular cameras to obtain binocular images;

[0053] The extraction unit 22 is used to extract the image information of the region where the pixel intensity gradient in the binocular image meets a certain threshold, and obtain the binocular image information;

[0054] Calculation unit 23, for calculating the map of the initial camera position through epipolar geometry according to binocular image information;

[0055] The construction unit 24 is used to calculate the current pose of the binocular camera by using the direct method and line features according to the information of the extracted front and rear frame images and the calculated prior three-dimensional map information during the movement of the binocular camera, and ...

Embodiment 3

[0058] The present embodiment provides a kind of binocular visual odometer based on direct method and line feature, described binocular visual odometer comprises: processor 31, memory 32; Described memory 32 stores code; Described processor 31 executes all Said code is used to implement the measuring method described in embodiment 1.

[0059] The binocular visual odometer based on the direct method and line features provided in this embodiment has the following advantages: it can provide calculation accuracy, and when large-scale movement and rotation occur, it can be tracked well and is not easy to be lost; in an environment with less texture It also has good robustness; and solves the scale uncertainty problem of monocular visual odometry; also reduces the initialization process of monocular visual odometry; and can make image matching more robust.

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 invention relates to the technical field of robot positioning and navigation, in particular to a binocular vision odometer based on a direct method and a linear feature and a measurement method. The measurement method comprises the steps of acquiring an image through a binocular camera at the same time to obtain a binocular image; extracting image information of a region witha pixel strength gradient meeting a certain threshold value in the binocular image to obtain binocular image information; calculating a map of an initial camera position through epipolar geometry according to the binocular image information; during the movement process of the binocular camera, calculating a current posture of the binocular camera by utilizing the direct method and the linear feature according to the extracted front and back frame image information and the calculated transcendental three-dimensional map information, positioning, and building the map according to the posture ofthe binocular camera at the same time. The binocular vision odometer based on the direct method and the linear feature and the measurement method provided by the embodiment of the invention has the advantages that calculation accuracy can be provided, and a robot can be well tracked and is not easy to lose when large-scale movement and rotation occur.

Description

technical field [0001] Embodiments of the present invention relate to the technical field of robot positioning and navigation, in particular to a binocular vision odometer and measurement method based on direct methods and line features. Background technique [0002] With the continuous development of robot technology, the traditional method of building an environment map and determining the pose of a robot is difficult to meet the requirements of robot computing speed and accuracy. [0003] At present, there are two main types of simultaneous localization and mapping (SLAM) for robots: SLAM based on feature point method and SLAM based on direct method. [0004] SLAM based on feature point method includes Parallel Tracking and Mapping (Parallel Tracking And Mapping, PTAM) and ORB (Oriented FAST and Rotated BRIEF, FAST with direction and Rotated BRIEF)-SLAM, etc., need to extract feature points and calculate descriptors, And to match the descriptors, the amount of calculatio...

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): G01C11/02G01C22/00
CPCG01C11/02G01C22/00
Inventor 高洪臣
Owner 轻客小觅机器人科技(成都)有限公司
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