Unlock instant, AI-driven research and patent intelligence for your innovation.

Dynamic feature point identification method and application thereof

A recognition method and dynamic feature technology, applied in image data processing, instruments, calculations, etc., can solve the problems of limited use scenarios, general processing efficiency, and no distinction effect, and achieve the effect of improving positioning accuracy and solving motion conflicts

Pending Publication Date: 2021-06-04
AEROSPACE INFORMATION RES INST CAS
View PDF5 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, this method performs generally in terms of processing efficiency, and has no distinguishing effect on dynamic objects other than the standard training set, and the usage scenarios are limited.

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
  • Dynamic feature point identification method and application thereof
  • Dynamic feature point identification method and application thereof
  • Dynamic feature point identification method and application thereof

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0043] The current robot positioning and mapping algorithms are mostly based on the assumption of a static world. When there are dynamic objects in the camera's field of view, the accuracy will decrease or even the positioning will fail. The invention proposes a dynamic feature point recognition method that combines computer vision and inertial measurement unit information. First, by pre-integrating the inertial measurement unit data, the relative pose change between the two frames of images is obtained, and then the motion model between the two frames of images is obtained by random sampling consistency algorithm, and compared with the pre-integration value of the inertial measurement unit , eliminate dynamic feature points, select static feature points to publish, and participate in back-end optimization to obtain more accurate motion trajectories. The method of the invention can quickly and effectively identify dynamic feature points, improve positioning accuracy in dynamic...

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 provides a dynamic feature point identification method, which is applied to the field of robot positioning and mapping algorithms, and comprises the following steps: acquiring an image, obtaining data of an inertial measurement unit sensor, and pre-integrating the data; performing alignment operation on the image and the difference value of the inertial measurement unit, and obtaining a pose change q_IMU sensed by the inertial measurement unit; carrying out feature point detection on the aligned front and back frames of images by adopting a feature point extraction algorithm, and carrying out feature point matching; obtaining a motion model R_camera and an inner point set inlier satisfying all feature point matching pairs between the front and rear frames of images through an RNASAC algorithm; comparing the q_IMU and the R_camera, and publishing feature points; and using the published feature points to carry out back-end optimization to obtain the moving trajectory of the camera.

Description

technical field [0001] The invention relates to the field of robot positioning and mapping algorithms, in particular to a dynamic feature point recognition method and its application that integrates computer vision and inertial measurement unit information. Background technique [0002] In recent years, with the continuous development of robot technology, the demand for autonomous positioning and movement of robots in the environment has become more and more urgent. Therefore, robot localization and mapping (Simultaneous Localization And Mapping, SLAM) technology has emerged as the times require. As the front end of SLAM, Visual Inertial Odometry (VIO) combines information from Visual Odometry (VO) and Inertial Measurement Unit (IMU) to estimate the pose of the camera. The main function of the visual odometry is to estimate the pose of the camera between adjacent images based on the feature points or pixels of the images captured by the camera. The main function of the inert...

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): G06T7/246
CPCG06T7/248G06T7/251G06T2207/30241G06T2207/30244
Inventor 蔡浩原李文宽赵晟霖柳雅倩刘春秀
Owner AEROSPACE INFORMATION RES INST CAS