Recognition of pipeline geometry by using monocular camera and PSD sensors

This study proposes a method of sensing basic pipeline elements, such as a straight pipeline, elbow, T-branch, and miter, by using a monocular camera and position sensitive device sensors. The method is composed of the three following parts: The pipeline elements are first determined; the T-branch a...

Full description

Saved in:
Bibliographic Details
Published in:Intelligent service robotics 2017-07, Vol.10 (3), p.213-227
Main Authors: Choi, Yun Seok, Kim, Ho Moon, Mun, Hyeong Min, Lee, Yoon Geon, Choi, Hyouk Ryeol
Format: Article
Language:English
Subjects:
Algorithms
Artificial Intelligence
Cameras
Control
Dynamical Systems
Elbow
Engineering
Geometry
Global positioning systems
GPS
Lasers
Mechatronics
Original Research Paper
Position sensing
Robotics
Robotics and Automation
Robots
Sensors
User Interfaces and Human Computer Interaction
Vibration
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This study proposes a method of sensing basic pipeline elements, such as a straight pipeline, elbow, T-branch, and miter, by using a monocular camera and position sensitive device sensors. The method is composed of the three following parts: The pipeline elements are first determined; the T-branch and miter are then classified among them; and the opening directions of the T-branch and the elbow are recognized. We develop a sensor hardware and signal-processing algorithm for providing information on the pipeline elements required to navigate inside the pipelines. This algorithm is easily implementable without any heavy computational burden. The proposed method is tested in an in-pipe robot, called MRINSPECT VI. Subsequently, its effectiveness is validated.
ISSN:1861-2776
1861-2784
DOI:10.1007/s11370-017-0221-1