Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A Vision Slam Algorithm Using CNN Feature Detection in Full Cycle

A feature detection, full-cycle technology, applied in computing, computer parts, instruments, etc., can solve problems such as inability to build trajectories and maps, unreliable results, etc., achieve concise visual odometry, fully express image information, and improve accuracy. Effect

Active Publication Date: 2022-07-01
BEIHANG UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Errors will inevitably accumulate over time, causing cumulative errors in the entire system, and long-term estimation results will be unreliable. In other words, we cannot build globally consistent trajectories and maps

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
  • A Vision Slam Algorithm Using CNN Feature Detection in Full Cycle
  • A Vision Slam Algorithm Using CNN Feature Detection in Full Cycle
  • A Vision Slam Algorithm Using CNN Feature Detection in Full Cycle

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0020] The present invention will be further described in detail below in conjunction with the accompanying drawings.

[0021] The present invention adopts the visual SLAM algorithm of CNNs feature detection in the whole cycle, such as figure 1 shown, including the following steps:

[0022] Step 1: Scan the surrounding environment information;

[0023] The binocular camera is used to move along the square area to collect the environmental image information of the real scene, and the obtained video stream is transmitted to the host computer in real time. The number of revolutions of the binocular camera is 1 to 2, forming a closed loop, which facilitates the compensation of the accumulated error in the subsequent closed-loop detection link. The above process is repeated, and a part of the multiple video streams collected is used as a training data set and a part is used as a test data set.

[0024] Step 2: Pre-training the training data set in the video stream collected in s...

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 SLAM algorithm that adopts CNNs feature detection in the whole cycle. First, at the front end, the original image data is pre-trained with an unsupervised model, and then the pre-trained data is used to combine the joint representation of motion and depth with the CNN network architecture. Local velocities and orientation changes are correlated to perform visual odometry; finally, path prediction is performed. The present invention also uses the OverFeat neural network model to perform the loopback detection link, which is used to eliminate the accumulated error brought by the front end and construct a visual slam architecture based on deep learning. At the same time, time and space continuity filters are constructed to verify the matching results, improve the matching accuracy and eliminate false matching. The present invention has great advantages and potentials in improving the accuracy of the visual odometer and the correct rate of closed-loop detection.

Description

technical field [0001] The invention belongs to the technical field of simultaneous localization and map construction (slam) algorithms in computer vision, and in particular relates to a visual SLAM algorithm that adopts CNNs feature detection in a full cycle. Background technique [0002] The Chinese name of SLAM (Simultaneous Localization and Mapping) is "simultaneous localization and map construction". SLAM is a fascinating area of ​​research with broad applications in robotics, navigation, and many other applications. Visual SLAM basically involves estimating camera motion and trying to build a map of the surrounding environment from visual sensor information, such as a sequence of frames from one or more cameras. The current research methods of SLAM problems are mainly to estimate the motion information of the robot body and the feature information of the unknown environment by installing multiple types of sensors on the robot body, and use information fusion to achiev...

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 Patents(China)
IPC IPC(8): G01C21/32G06V10/74G06V10/764G06V10/774G06V10/82G06K9/62
CPCG01C21/32G06F18/22G06F18/241G06F18/214
Inventor 赵永嘉张宁雷小永戴树岭
Owner BEIHANG UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com