Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Matching-range-constrained real-time loop closure detection with CNNs features.

Dongdong Bai1, Chaoqun Wang1, Bo Zhang1

  • 1College of Computer, National University of Defense Technology, Changsha, China ; State Key Laboratory of High Performance Computing, National University of Defense Technology, Changsha, China.

Robotics and Biomimetics
|October 13, 2016
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

Difference from Background: Limit of Detection01:05

Difference from Background: Limit of Detection

8.7K
The limit of detection (LOD) is the smallest amount of analyte that can be distinguished from the background noise. The LOD value corresponds to the concentration at which the analyte signal is three times larger than the standard deviation of the blank signal. Below this value, the analyte signal cannot be differentiated from the background noise. It is calculated by dividing the calibration slope by 3 times the standard deviation of the blank signals.
The LOD indicates the presence or absence...
8.7K

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Reamed and unreamed intramedullary nailing for the treatment of open and closed tibial fractures: a subgroup analysis of randomised trials.

International orthopaedics·2009
Same author

Selective COX-2 inhibitor versus nonselective COX-1 and COX-2 inhibitor in the prevention of heterotopic ossification after total hip arthroplasty: a meta-analysis of randomised trials.

International orthopaedics·2009
Same author

[Study on evaluating sex determining region of the Y as an engrafting track of BMSCs transplantation for repairing osteonecrosis of the femoral head of rabbit].

Zhongguo xiu fu chong jian wai ke za zhi = Zhongguo xiufu chongjian waike zazhi = Chinese journal of reparative and reconstructive surgery·2009
Same author

Positive association between benign familial infantile convulsions and LGI4.

Brain & development·2009
Same author

Catalytic enantioselective synthesis of chiral phthalides by efficient reductive cyclization of 2-acylarylcarboxylates under aqueous transfer hydrogenation conditions.

Organic letters·2009
Same author

Significance of urinary liver-fatty acid-binding protein in cardiac catheterization in patients with coronary artery disease.

Internal medicine (Tokyo, Japan)·2009
Same journal

Implementation of Q learning and deep Q network for controlling a self balancing robot model.

Robotics and biomimetics·2019
Same journal

Cognition-based variable admittance control for active compliance in flexible manipulation of heavy objects with a power-assist robotic system.

Robotics and biomimetics·2018
Same journal

PID, BFO-optimized PID, and PD-FLC control of a two-wheeled machine with two-direction handling mechanism: a comparative study.

Robotics and biomimetics·2018
Same journal

Systematic engineering design helps creating new soft machines.

Robotics and biomimetics·2018
Same journal

Hybrid control combined with a voluntary biosignal to control a prosthetic hand.

Robotics and biomimetics·2018
Same journal

A multi-jointed underactuated robot hand with fluid-driven stretchable tubes.

Robotics and biomimetics·2018
See all related articles

This study enhances visual loop closure detection (LCD) for robots by using deep convolutional neural networks (CNNs). The new method improves accuracy and real-time performance in challenging robotic environments.

Area of Science:

  • Robotics
  • Computer Vision
  • Artificial Intelligence

Background:

  • Loop closure detection (LCD) is crucial for visual simultaneous localization and mapping (SLAM) systems to correct accumulated drift.
  • Deep convolutional neural networks (CNNs) show promise for visual LCD, but challenges exist in robotic applications due to image similarities and real-time demands.

Purpose of the Study:

  • To adapt CNN features for effective visual loop closure detection in real-world robotic environments.
  • To address challenges of high similarity between adjacent images and the need for real-time performance in robotic SLAM.

Main Methods:

  • Utilized features generated by CNN layers for visual loop closure detection.
  • Implemented a value-limited image matching range to differentiate loop-closing images from similar adjacent ones.
Keywords:
CNNsFeature compressionLoop closure detection

Related Experiment Videos

  • Employed an efficient feature compression technique to enhance real-time processing.

    • Main Results:

    • Achieved superior results compared to state-of-the-art methods in loop closure detection.
    • Significantly improved the real-time performance of the visual LCD system.
    • Successfully addressed the challenge of distinguishing loop closures from highly similar sequential images.

    Conclusions:

    • The proposed CNN-based approach effectively implements loop closure detection in real robotic environments.
    • The method offers a robust solution for accurate and efficient localization in SLAM systems.
    • This work advances the application of deep learning for real-time robotic perception tasks.